Boucher Ecology of Mutualism

Ang Rev. Ecol. Syst. 1982. 13:315--47Copyright © 1982 by Annual Reviews lnc. All rights reserved

THE ECOLOGY OFMUTUALISM

Douglas 1t. Boucher

Departement des sciences biologiques, Universit~ du Quebec ~ Montreal, C. P.8888, Suet. A, Montreal, Quebec, Canada H3C 3P8

Sam James

Department of Ecology and Evolutionary Biology, University of Michigan, AnnArbor, Michigan, USA 48109

Kathleen H. Keeler

School of Life Sciences, University of Nebraska, Lincoln, Nebraska, USA 68588

INTRODUCTION

Elementary ecology texts tell us that organisms interact in three fundamen-tal ways, generally given the names competition, predation, and mutualism.The third member has gotten short shrift (264), and even its name is notgenerally agreed on. Terms that may be considered synonyms, in whole orpart, are symbiosis, commensalism, cooperation, protocooperation, mutualaid, facilitation, reciprocal altruism, and entraide. We use the term mutual-ism, defined as "an interaction between species that is beneficial to both,"since it has both historical priority (311) and general currency. Symbiosisis "the living together of two organisms in close association," and modifiersare used to specify dependence on the interaction (facultative or obligate)and the range of species that can take part (oligophilic or polyphilic). Wemake the normal apologies concerning forcing continuous variation anddiverse interactions into simple dichotomous classifications, for these andall subsequent definitions.

Thus mutualism can be defined, in brief, as a -b/q- interaction, whilecompetition, predation, and eommensalism are respectively -/-, -/q-, and-t-/0. There remains, however, the question of how to define "benefit to the

3150066-4162/82/1120-0315 $02.00

Annual Reviewswww.annualreviews.org/aronline

Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.

http://www.annualreviews.org/aronline

316 BOUCHER, JAMES & KEELER

species" without evoking group selection. Two definitions have coexistedfor decades. In one, benefit is defined on the individual level, in terms of therelative fitness of organisms that do and do not participate in the interac-tion. This definition seems particularly appropriate for questions of evolu-tion, but is inadequate when considering population-level phenomenainvolving a balance of positive and negative fitnesses. For example, manyfrugivorous animals destroy some of the seeds they consume and disperseothers; the net result of these individual predations and mutualisms may beeither an increase or a decrease in the plant population. The second defini-tion considers this population-level effect, and is the definition tacitly as-sumed in applying models such as Lotka-Volterra equations to mutualism(as well as competition and predation). Thus the individual-level definitionuses change in the fitness, l/z I of individuals of species 1 when they interactwith individuals of species 2, while the population-level definition uses thechange in dN~/dt, the growth rate of population 1, as N2 changes. We willnot attempt to enshrine one or the other of these definitions.

We cover first direct mutualism, in which the two species interact physi-cally, and then indirect mutualism, in which each species benefits from theother’s presence but there is no direct contact. Direct mutualisms are di-vided into symbiotic and nonsymbiotic mutualism, using physiological inte-gration as the basic criterion. This approach is artificial but convenient. Ithas little general usefulness: other divisions of mutualism (289) do not alignneatly on a symbiotic/nonsymbiotic dichotomy. Although exceptionsabound, symbiotic mutualisms tend to be coevolved and obligate, whilefacultative mutualisms are frequently nonsymbiotic and not eoevolved.

Using these definitions, we start with two observations. On the one hand,an enormous number of ecologically and economically important interac-tions, found throughout the biosphere, would seem to be mutualistic. Onthe other hand, few studies have actually demonstrated increases in eitherfitness or population growth rate by both of the species in an interaction.Interactions have generally been shown to be mutualisms by describingwhat is exchanged. Mutualism may be everywhere, but its existence remainspractically unproven.

HISTORY

The history of the study of mutualism by ecologists is akin to SherlockHolmes’s case of the barking dog, in which the point of interest was thatthe dog did not in fact bark. It is notable that despite promising earlybeginnings and wide recognition of mutualistie interactions, ecologists havedevoted little time and energy to this subject (264).


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.


ECOLOGY OF MUTUALISM 317

Although discussions of the subject of mutualism typically start withDeBary’s book on symbiosis (69), apparently the first use of the term"mutualism" is in Pierre van Beneden’s 1875 book Les Commensaux et lesParasites (311), published simultaneously in German and English (as Ani-malParasites and Mess-Mates). This was a popular work, and van Benedenwas perhaps the foremost Belgian biologist of his time, with a bibliographyranging from worms to whales (172). In his youth he was active in therevolution of 1830, which won Belgian independence. He became celebratedin later years for his parasitological work, on which Les Commensaux et lesParasites concentrates. But he felt that since, in addition to parasites, "wefind others who mutually provide each other services, it would be mostunflattering to call them all parasites or commensals. We consider it fairerto call them Mutualists, and thus mutualism takes its place beside com-mensalism and parasitism" [(27); italics in original]. The capitalization "Mutualists" is probably an indirect reference to the "Mutualitr" societiesorganized by workers in France and Belgium to support each other finan-cially, and indeed analogies to human society are common throughout thebook.

Only two years later, van Beneden’s definitions were critically discussedby Alfred Espinas in a doctoral thesis at the University of Paris entitledDesSoci~t~sAnimales. This work, which shocked the university authoritiesby discussing the philosophy of Auguste Comte, "whom no one at that timedared to mention" (80), is concerned with the question "what is the essenceof society?" Espinas discusses mutualism in a chapter on "Accidental soci-eties between animals of different species: Parasites, Commensals, Mutual-ists." The treatment concentrates on domestication as one kind ofmutualism, but also discusses such relationships as tick-birds and rhinocer-oses, mixed species bird flocks, and ant-aphid associations.

By 1893 the subject was sufficiently developed to deserve a review articlein the American Naturalist by Roscoe Pound (250). Pound recognized mostof the major kinds of mutualism we study today, including pollination,legume root-nodulation, and various animal examples. By the turn of thecentury one could count literally hundreds of articles on one mutualism oranother (273). More importantly, there seems to have been a general recog-nition of the fundamental similarity of interactions ranging from mycor-rhizae to cleaner-fish.

The analogy of mutualism to cooperation in human society, never absentin previous work, was made a political issue in 1902 with the publicationof Mutual Aid: A Factor in Evolution, by the anarchist Peter Kropotkin(178). This best-selling work was a rebuttal of the Social Darwinists andcited examples of cooperation in the natural world, to counter the conten-tion that nature proved the inevitability of cutthroat competition. With


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



Kropotkin’s book, mutualism became well known in lay society, and itcontinued to be discussed among biologists in the following decades. Butin comparison to the rapid growth of the rest of ecology in the 20th century(76, 340), mutualism lost ground. While the construction of a theoreticalbase for studies of competition and predation in the 1920s and 1930s gavethese subjects a strong boost, (93, 94, 194, 277, 318) the theoretical workof Kostitzin on mutualism (177) was amost totally ignored.

Kostitzin led a fascinating life (276, 277): After fighting in the RussianRevolutions of 1905 and 1917, he became head of the Soviet Union’sGeophysics Institute. In the mid-1920s he moved to France, where he livedtill his death in 1963. He collaborated with ¥ito Volterra, and published ontopics ranging from single-species population dynamics to glaciation andthe evolution of the atmosphere.

Mutualism figures prominently in the work of Warder C. Alice, a Quakerand pacifist, who wrote extensively on human and animal cooperation andcoauthored the textbook Principles of.4nimal Ecology (6). This work wasfor many years the fundamental text in animal ecology, and it devotesconsiderably more space to mutualism than most contemporary ecologybooks. Yet it revealed the lack of progress made since the 19th century.Allee et al’s collection of examples is longer than Pound’s, their presenta-tion of the analogy between natural mutualisms and cooperation in humansociety is subtler than Kropotkin’s, but the treatment contains little that isnew.

Only in the 1970s has mutualism finally begun to compete with otherinteractions as a subject for ecologists’ consideration. To judge by citations,their work has relied little on that of their predecessors; during the period1965-1979, van Beneden’s book was cited almost exclusively by parasitolo-gists, Kropotkin’s mostly by social scientists and sociobiologists (oftenunfavorably), and Pound’s article hardly at all (145). Interestingly, severalrecent theoretical treatments of mutualism have depicted it by means of thesame graph, though with little cross-referencing (41, 84, 139, 213, 288, 315,316). Mutualism’s time seems finally to have arrived.

Two points emerge from this brief historical survey: the lack of interestin mutualism among ecologists for most of the 20th century, and the in-volvement of many of those who did study it with what at the time wereleft-wing causes. We suggest as an hypothesis for historians of science thatmutualism has been avoided during most of the 20th century because of itsassociation with left-wing politics (perhaps especially with Kropotkin).

DIRECT MUTUALISM

Mutualisms have long been seen as exchanges of benefits, of which one canidentify a few main types: (a) nutritional: either the breakdown of com-pounds by digestion for the partner, or supply of growth factors or nutrients


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



by synthesis or concentration; (b) supply of energy, generally from photo-synthesis; (¢) protection, either from environmental variation or from ene-mies; and (d) transport, either from unsuitable to suitable environments by dispersal of gametes or propagules. Symbiotic mutualists generally ex-change the first two services, sometimes the third, and rarely the fourth.Nonsymbiotic mutualisms can involve all four.

Symbiotic Mutualisms

TYPES Most symbiotic mutualisms involve the supply of energy from onepartner to the other, whether autotroph to heterotroph (284) or betweenheterotrophs (e.g. gut microbes). A number of benefits may be provided exchange: (a) breakdown of compounds, facilitating digestion; (b) supplyor concentration of nutrients; (c) environmental constancy; and (d) biolu-minescence.

Gut flora are involved in breaking down cellulose and related substancesin mutualism with many vertebrates (134, 136), as well as with termites(304) and other arthropods (51, 291). Urea is broken down and its nitrogenrecycled by rumen bacteria (134) and by the fungal components of somelichens (4). Toxic secondary plant compounds are also degraded in caecaand rumens by microbial symbionts (134).

The synthesis of compounds, including vitamins and amino acids, isperformed by gut microbes in numerous associations (12, 48, 51, 77, 123,134, 174). Ruminant microbes, and those in other animals, synthesize allof the B-complex vitamins, vitamin K, and the sulfur-containing aminoacids. The breakdown of urea results in ammonia and carbon dioxide, andthe ammonia is used by the bacteria to make amino acids, which are laterobtained by the animal (134). Similar activities take place in lichens which ammonia is incorporated by algae into amino acids and vitamins, andvitamins are secreted by the algae (4). An endozoic alga has been found be the major source of fatty acids to its host flatworm, in a case in whichthe flatworm has lost the ability to synthesize these substances (215). My-corrhizal fungi also produce growth factors, in this case plant hormones(42).

Symbiotic nitrogen fixation is known to take place in root nodules ofleguminous plants (31, 71, 228, 256, 339) and in some nonlegumes (38, 267, 303). In the legumes, the bacteria of the genus Rhizobium are theendosymbionts, and in the others, Actinomycetes, except for a recentlydiscovered Rhizobium-Ulmaceae association (23, 308, 309). Blue-green al-gae fix nitrogen in both roots and above-ground parts of other woody plants,such as cycads, Gunnera, and various Rubiaceae (256). Symbiotic nitrogenfixation has also been discovered in lichens whose algal components areblue-green algae (4, 216), sphagnum moss (107), aquatic ferns (197), grasses


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



(43, 227), sponges (333), sea urchins (112), shipworms (55), and (25, 44). A disadvantage is that the cost of nitrogen to a plant with sym-bionts is greater than to a plant that acquires nitrogen in the form of nitrate(113).

Mutualisms involving the concentrating of nutrients include mycorrhizaeand a few other instances involving algae and bacteria. Mycorrhizal fungican take up nutrients at low concentrations, improve water uptake, andconfer resistance to pathogens upon their hosts (42, 62, 87, 146, 147, 257).Plants benefit most when soil nutrient concentrations are low. In somecircumstances, normally mutualistic fungi can be parasitic, and in rareinstances of high phosphate levels the increased uptake caused by mycor-rhizae can cause phosphate poisoning (42). Nutrient uptake by endozoicalgae has been found in coral zooxanthellae, which take up ammonia fromseawater (222), and in green hydra, whose bacterial symbionts increaseuptake of phosphate from the medium (332). Endozoic algae in general takeup waste products from their hosts and use them for their own growth. Incorals, nitrogenous wastes have been traced from zooplankton to polyp toalgae, in what constitutes a recycling of nutrients within the association(224, 248). Leaf-cutting ants (Attini), macrotermitine termites, and a vari-ety of beetles culture fungi (20, 21, 92, 255, 327, 337). They feed, protect,and distribute the fungus, consuming parts of it (which are often specializedfor the purpose) in return. The fungi seem in many cases to be incapableof independent existence. These interactions are best considered symbioticeven though the fungi are external to the ants and physically encounteredby them only during culture: They share the colony’s nest, and the bio-chemical and behavioral interactions are complex (255, 337).

Green plants are also "fed" by ants that inhabit the plant and bring debrisinto their nests, which decomposes and is available for uptake by the plant(140, 141, 146, 265). In "ant-gardens," the ants place the plants’ seeds their nests, where they germinate and take up nutrients from the nest wall(173). In at least one species, when the ants are absent the plant stopsproducing the food bodies on which the ants feed (266).

The provision of a constant environment, or a place in which to lead asheltered existence, is found in mutualisms involving endozoic algae and inlichens. Endozoic algae generally have thinner cell walls than free-livingalgae, a characteristic that may be due to the physical protection gained byliving within another organism (201, 294). Similarly, the algae found lichens are frequently unable to survive where the lichen is found growing,suggesting that living with fungi improves resistance to injury and desicca-tion (4). There are trade-offs, however; symbiotic species of algae grow moreslowly when inside animals than when cultured in vitro, and more slowlystill than nonsymbiotic species (324).


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



Symbioses with bioluminescent microorganisms are found in a variety ofanimals (126, 207, 226). While bioluminescence may have originated as oxygen detoxification system (207), it now has a wide variety of functions,including prey attraction, protection, communication, and mate attraction.In many eases the function is not yet understood. This is thus a "grab-bag"category in which many different benefits share a common biochemistry.

EVOLUTION The evolution of symbiotic mutualisms is generally thoughtto begin through proximity of the organisms involved. We discuss howassociations are initiated, what sorts of adjustments and adaptations mustbe made en route to mutualism, and the selective pressures for thesechanges.

Parasitism is one possible starting point (271,274, 327). In this case theproblem of first infection has beentaken care of, and one needs only toconsider how parasitism could change to mutualism. A model system is theexperimental demonstration of reduction of virulence of a bacterial infec-tion of an Amoeba, which ultimately became dependent on the bacteria(160). Reduction of detrimental effects of the parasite on the host must accompanied by development of host dependency on the parasite, which isalready dependent on the host. Scott (274, 275) gives two routes to theevolution of dependence. The first is through a parasite-relationship of lowvirulence in which the parasite leaks nutrients, increasing host survival andthus its own fitness. Alternatively, an increase in ecologically imposedlimitations, which the parasite helps overcome, can select for closer depen-dence.

Ingestion as a starting point is thought to be appropriate for mutualismsinvolving algae, gut microbes, and some eukaryotic organdies (205, 206,207). Selection pressure for resistance to digestion would be strong, and theingesting animal would have to develop a means of recognizing the alga asa nonfood item, and also as nonforeign. Carbohydrate diffusion from thealga would place some value on retention of the intact alga. Another neces-sary alteration is the regulation of growth rates of the algae (223). Currentlyalgae are found in vacuoles or in the cytoplasm, and some are reduced tolittle more than chloroplasts (283, 325).

In cases starting as symbiotic commensalisms, the evolution of mutual-ism may proceed by the commensal’s providing some benefit that would beselected for if it increased the host’s chances of survival and/or decreasedthe likelihood of the host’s attempting to get rid of the commensal. Alterna-tively, a change in ecological circumstances or the presence of an ecologicalopportunity could transform the relationship into a mutualism if the eom-mensal happens to render the host better able to survive or take advantageof the situation.


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



A driving force behind the evolution of symbiotic mutualisms is ex-pressed by Dubos & Kessler (73), who argue that as more needs of theassociation are met by the combined abilities of the mutualists, the intensityof competition on those partners from ecologically similar species willdiminish. This proceeds by selection of one symbiont to increase its fitnessthrough changes that increase the fitness of the other symbiont.

Nonsymbiotic MutualismsNonsymbiotie mutualisms are those in which the two species are physicallyunconnected. While there may be some physiological habituation to themutualist, as in the anemone-dwelling fishes (89), there are no direct.physio-logical links. Various mutualisms involving ants, including myrmecophytesand the leaf-cutter ant-fungus relationship, offer ample material for argu-ments about classification; having considered nutritional aspects above, wedeal with protection in this section.

TRANSPORT Pollination and dispersal have enjoyed a recent surge ofinterest and have accumulated a large literature to which we cannot dojustice in the scope of this paper (but see reviews by Howe, Regal, andDressier in this volume). Both involve the transport of particles involvedin reproduction, and for both there are nonmutualistic alternatives. In bothinteractions, transport is effected in exchange either for some sugar-richsubstance (nectar, ovarian tissue of fruits) or the consumption of some the particles to be transported (pollen, nuts). In the case of pollination, which the "target" area is very limited, the transport will generally be eithersuccessful or totally unsuccessful. In dispersal, on the other hand, thedifference between outcomes is not as absolute, and the target is large anddiffuse (328). Finally, both interactions are important to the genetic struc-ture of the population.

Pollination by mutualism with animals is critical to sexual reproductionin the majority of flowering plants. The advantage of the mutualistic solu-tion may be reduced pollen waste, longer transport, or increased probabilityof success at low density (290). After initiation with insect species in theColeoptera and Diptera, the mutualism has been developed in several otherorders, most notably the Hymenoptera (110, 127, 129, 158, 252, 297). Thereis a variety of vertebrates involved, including birds, bats, monkeys (148),lemurs (293), rodents (331) and other mammals.

Dispersal of seeds, fungal spores (19, 88, 95, 210), and other propagulesmay not be as critical as pollination;.offspring could simply grow up beneaththeir parent and replace it. However, dispersal would appear to offer the(seldom proven) advantages of colonization of Other sites, increased out-crossing, and escape from predators (124, 125, 231, 243). Since some


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



those predators may be the dispersers themselves, the interaction can shiftbetween mutualism and predation depending on plant and predator/dis-perser densities (39). Many animal taxa act as dispersers, includingbirds, (135,200), bats (85), large mammals (159), ants (22, 30, 66, 119, 125, 133, 195, 231,253, 298), Drosophila (95), beetles (19), tortoises (262),and fish (105).

PROTECTION Protection from predators, parasites, diseases, toxins(161), and occasionally competitors is provided by many organisms. Theother partner may be animal or plant, and may provide food (sensu lato),reciprocal protection, housing, or some combination.

Ant-plant protective mutualisms range from symbiotic obligate systemsin which the plants house and feed the ants [e.g. Acacia (24, 149); Cecropia(150)] to housing ants without food provided [e.g. rubiaceous myrmeco-phytes (29, 140, 141); Barteria (154)] and food provided without nest sites[chiefly extrafloral nectar-producing species (27, 82, 342)]. In all cases,aggressive and predatory ant behavior serves to reduce damage to the plant(26, 27, 28, 35, 70, 132, 141, 143, 149, 150, 154, 163, 165, 167, 168, 171,176, 229, 239).

Ants and other insects are also mutualists with some herbivorous andsap-feeding insects, the ants providing protection in return for honeydew,a sugary secretion extruded through the anus or glands (1, 2, 9, 46, 130, 241,270, 323). Homoptera and Lepidoptera seem to be the main taxa involved.Related mutualisms are some cases of phoresy in which insects transportother insects or mites on their backs. The "riders" are brought to sourcesof food and in return may protect the insects against parasites (336). Similarbenefits may occur with nest-sharing insects (337).

There are numerous examples of protection mutualisms among marineanimals, including burrow-sharing by gobies and shrimp (89, 111, 192),anemones living on crab shells (15, 269), and cleaning mutualisms (77, 106, 131,188, 190, 191,193, 282). Both the gobY-shrimp and crab-anemonemutualisms allow the partners to occupy areas they otherwise could not (89,269). Cleaning interactions are a set of potential mutualisms that have beena matter of some controversy (131, 190, 191, 193, 282). Cleaners are knownfrom a variety of marine habitats, and there are many species apparentlyadapted to the role. Their removal only sometimes has negative effects onthe host fishes. Losey (193) questions the mutualistie nature of some clean-ing ¯ relationships, noting that parasite load shows little correlation to ten-dency to solicit cleaning, and that scales and mucus are also taken bycleaner fishes.

Other interactions involve mutual protection, which may appear as asimple consequence of living together. Protection against starfish predation


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



is provided to corals, scallops, and clams by xanthid crabs, sponges, anddiverse epibionts, respectively (36, 102, 313). The epibionts and sponges arein turn less vulnerable to predation, and the crabs gain both protection andenergy-rich mucus. Further benefits are provided in the mutualism betweensea-anemones and the fish (Amphiprion) that live inside them (89, 208). Theanemone gains protection, food, and is cleansed of waste and necrotictissues, while the fish is protected and suffers less parasitism and disease.In other invertebrates, such protection mutualisms can help species incompetition for space as well (52, 232). This aid can allow both partnersto extend their habitat ranges, and also, for example, permits the bryozoanCelleporia to dominate the community, which it does not do in the absenceof the mutualistie hydroid Zanclea (232).

Both increased food and protection may be involved in mixed-speciesaggregations; but, which is more important, and indeed the mutualisticcharacter of the phenomenon itself, are controversial. Mixed aggregation isknown in delphinid whales (61,245), fishes (192), birds (58, 90, 217, 221), sea urchins (74) and terrestrial mammals (11, 32). The two principalhypotheses relr-ting to the formation of heterospecific groups are that theyenjoy an increased efficiency of foraging and that they are better able to todetect and avoid predators. Bird flocks have been the most thoroughlystudied.

Increased foraging etficiency by flocking has been proposed by Moynihan(220) and Cody (59). Flocking tends to occur when food availability is (59, 217). However, individual birds in flocks spend less time watching andmore time feeding, suggesting that protection from predators is the primaryreason for ttocking, and improved feeding is a secondary consequence (104,181). The theories of Vine (317) and Hamilton (118) show that individualscan reduce their chances of being eaten if they cluster together. This helpsexplain the origin of flocking for predator avoidance but not its specificallymutuaIistic (multi-species) nature. The motivation for mixed grouping vaguer for monkeys than for birds. Mixed troops of monkeys travel and feedtogether (32), and the different species recognize one another’s alarm calls(209). Baboons and impalas or bushbucks benefit from each other’spresenceby mutual recognition of alarm calls (ll, 78, 321). In association, theimpalas enter types of vegetation they do not use when unaccompanied bybaboons (78).

APPARENT PREDATION A third class of nonsymbiotic mutualism,somewhat miscellaneous and in many cases speculative, consists of appar-ently predatory interactions in which the "prey" actually benefits. Porter(247) has shown that some algae absorb phosphorus from the gut of zoo-plankton as they are "eaten" and usually pass through unharmed; similar


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



benefits may occur in copepods (79). Enhanced availability of nutrients plants upon grazing has been suggested by Owen & Wiegert (235, 236) andby Stenseth (292), while growth-enhancing factors in grasshopper and bisonsaliva have been studied with mixed results (75, 128). Grazers may alsoretard succession and thus preserve the plants of the successional stagesthey prefer (56, 301).

Other predators may reduce allocation of resources to offspring unlikelyto survive and thus make their prey more efficient in reproduction (310).Similarly, herbivores may change the growth patterns of their prey andpossibly decrease vulnerability to physical damage (281). As these casesindicate, what is now a mutualism may have arisen from some adaptationof the prey to predation, which makes it inefficient when the predator isabsent. But whatever the origin, the interaction must be seen as mutualisticif fitness and/or population growth are reduced without it. The same princi-ples apply to nontrophic interactions that appear to be +/-, such as cow-birds’ laying eggs in other birds’ nests or epiphytes weighing down trees’branches: The negative impact may be exceeded by benefits such as parasiteremoval (285, 286) or nitrogen fixation and trapping of nutrients the treecan use (225). These cases, while often speculative, are fundamentally simi-lar to those of pollen-feeding pollinators or nut-consuming dispersers: Thereis a balance of positive and negative effects on the "prey" that may shift withenvironmental conditions (39, 341).

THEORIES OF MUTUALISM

Ecological theory about mutualism, excluding Kostitzin’s book (177), datesonly from the last decade, and has been directed at two main questions:(a) When will mutualisms develop (note that we do not say "evolve") in what sorts of species and environments will they be found? (b) When willa community involving mutualists persist (again, we do not say "be sta-ble")?

Excluding theories concerning indirect mutualism, to be dealt with in alater section, four kinds of model can be distinguished: (a) those of individ-ual selection, which are often of a cost-benefit type; (b) population dynamicsmodels, with two, three, or many species; (c) models of shifts of interactionsfrom mutualistic to predatbry or competitive; and finally (d) the "keystonemutualist" concept.

Individual selection models (13, 170, 249, 271, 335, 336), though differentin details, come to similar conclusions. They generally find that intimacyor symbiosis favors mutualism because the number of competitors receivingbenefits is restricted. However, Wilson’s (336) model shows that mutualismcan evolve even with large trait groups. Roughgarden (271) and Keeler


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



(170) each find that major fitness gains are critical. Axelrod & Hamilton(13) do not consider games with differing payoffs (the equivalent of changesin fitness), but it seems clear that even within the "prisoner’s dilemma"payoff matrix the greater the reward for mutual cooperation relative tononcooperation the more likely is cooperation. Mutualism is expected whenit is valuable [e.g. protection in areas of high predation pressure (15),nodulation under competition for soil nitrogen (40)] and when it is cheapand efficient [e.g. extrafloral nectaries when ants and sunlight are abundant(170)]. What is still lacking is a theory that will predict when mutualistic

¯ solutions are preferred to nonmutualistic ones, assuming both are possible(but see 113, 260).

The population dynamics models, most starting from Lotka-Volterracompetition theory and reversing the sign of the coefficients, are to someextent replies to the early contention that mutualism has a destabilizingeffect on communities (211, 212, 213) and therefore should be rare. It hasbeen shown that mutualism can be stabilized by a variety of means: strongnegative density-dependence (103, 116, 183, 249, 315), curvilinearities (3,116, 213, 288, 315, 329), frequency dependence (183), or predation (125).Furthermore, mutualism demonstrates situations in which models withouteither stable or feasible equilibrium points nevertheless have both speciespersisting indefinitely (315). Thus the criteria for stability, though widelyused, are probably irrelevant to existence in the real world (315), and evenhighly unstable models with both mutualism and positive density-depend-ence may be biologically reasonable (D. H. Boucher, in preparation). Lotka-Volterra models also indicate that mutualisms with stable equilibria tendto show out-of-phase oscillations (100).

Post, Travis & DeAngelis have shown how the mathematical theory ofM-matrixes can be used to examine communities in which all the interac-tions are mutualistic (68, 249, 305, 306). This theory may prove useful--even though communities of even three species in which all interactions aremutualistic are probably rare--since the theory also applies to communitieswith certain combinations of mutualistic and competitive interaction.

Other theories of mutualism are less mathematical. Boucher (39) hasdeveloped a graphic model of seed consumption to predict the populationdensities at which mutualism will turn into predation; the model success-fully predicts the relative densities of nut-producing trees. Gilbert’s conceptof a "keystone mutualist" (98) whose demise would produce major shiftsin community structure is analogous to the earlier "keystone predator" ideaof Paine (237).

Indeed, one approach to generating predictions about the effect of mutu-alism would be simply to reverse predictions concerning predators or com-petitors. For example, if removal of a predator on a competitively dominant


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



species should decrease diversity, removal of its mutualist should increaseit. Mutualism theory is still at a stage in which simple modifications ofcompetition and predation models may provide unexpected insights.

We can also suggest several questions about mutualism that theoristshave not considered. When will mutualisms tend to be asymmetric orsymmetric (46)? If one mutualist can interact with each of two others, whenwill the two "compete" for it, and how? When will the fitness and popula-tion-dynamics definitions of mutualisms conflict (288)? Finally, theory thatgives quantitative predictions about mutualism is almost totally lacking.Theoretically, the field is wide open.

THE DISTRIBUTION OF MUTUALISM

There are clear differences among communities in the abundances of somemutualisms. For example, angiosperms of tropical rainforests and tropicaldeciduous forests are almost entirely animal-pollinated (162), while thedominant species of temperate deciduous forests are wind-pollinated (=nonmutualist). Desert and tundra have numerous animal-pollinated species(81), while temperate grasslands, frequently 90% graminoids by canopycoverage, have little animal pollination. Boreal forests are similarly domi-nated by wind-pollinated trees. Comparable patterns hold for seed dis-persal.

Myrmecophytes--plants inhabited by ants--appear to be confined to thetropics (29, 141, 149, 150). Janzen (149) found ant-acacias to be limited the length of the dry season; a healthy ant colony required continuous leafproduction by the plant. There are other peculiar patterns: Both ants andacacias are numerous and important in Australia, but ant-inhabited acaciasdo not occur there (132).

Ant-plant mutualism at extrafloral nectaries, while having many of thesame properties, is much less specialized and more widespread. Both tem-perate and tropical plants are known to possess extrafloral nectaries (27,166, 342). Existing data are incomplete but suggest that the majority of suchspecies are tropical. Keeler (170) has recently summarized the data on coverby plants with extrafloral nectaries; it ranges from nearly 100% to zero buta temperate-tropical comparison would be premature. Most studies (27,164, 166, 170) have found positive correlations between abundance of extra-floral nectaries and ant abundance at the site.

Ant dispersal of seeds is widespread, but some regions [e.g. Australia (30,66)] are particularly rich. Beattie & Culver (22) found a significant correla-tion between the number of species with ant-dispered seeds and ant abun-dance at West Virginia sites, but cover by myrmecochorous plants wasindependent of ant abundance. Although myrmecochory is best known in


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



temperate plants, several tropical myrmecochores have recently been re-ported (133, 195), suggesting that current biogeographic information incomplete.

Similar temperate-tropical differences show up for other mutualisms.Cleaner-fish mutualisms, fungus-gardening invertebrates, and symbioticcorals seem to be confined to the tropics. Nitrogen-fixing mutualisms, onthe other hand, seem to contradict the pattern, with actinomycete-nodulated plants being rather boreal and the most tropical of the legumesubfamilies (Caesalpinoideae) having only about one fourth of its generaknown to nodulate. The proportions for the more temperate Papilionoidaeand Mimosoideae are about 90% (7).

Mutualisms thus seem more prominent in tropical communities (213),although there are some major exceptions. However, there are several prob-lems in interpreting this trend ecologically. There are major phylogeneticinfluences, and the mutualisms are by no means independent. [To addanother example, Gilbert (97) notes the inverse correlation of ant-plant andant-insect protection mutualisms.] Most fundamentally, the data in mostcases are absolute abundances and not proportions. While mutualism in-creases toward the equator, so do species richness, productivity, biomass,and perhaps predation pressure. Thus without data on proportions of taxaor individuals that are mutualistic, we can say little.

Two other claims have been made concerning the distribution of mutu-alisms: that they require environmental stability (91, 261), and that theyallow survival in marginal habitats (187). The apparent contradiction somewhat reconciled by the fact that the first claim applies more to nonsym-biotic mutualisms and the second to symbioses. It is true that many plantcolonists of disturbed areas do not form relationships with myeorrhizalfungi (146, 147), and many weedy plants are selfing or apomictic (122).However, light-gap species are frequently mutualistic (26, 230, 299).

It has also been suggested that mutualisms will allow survival in marginalhabitats. The nutritional aspect is of greatest ecological importance here,with the mutualistic association better able to survive in nutrient-poorhabitats (187) and on lower quality diets than the same species livingseparately. This can be thought of as an example of division of labor, withthe elimination of redundant metabolic pathways within the association.The association is like a microcosm, having its own nutrient cycling mecha-nisms, which tend to keep essentials within the system rather than let themflow through in an unrestricted fashion. Consequently, the mutualists areable to thrive in circumstances, such as the nutrient-poor waters of tropicaland subtropical seas or the poor soils of tropical rain forests, where nutrientloss is a severe problem (8).


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



Whatever the patterns of distribution, there is no doubt that mutualismsare important to the organization of many communities and the diversityof many taxa. Prime examples are the angiosperms, whose diversity anddominance expanded in tandem with those of their pollinators and, to alesser extent, their dispersers (50, 108, 242, 259, 290). The result has beena hundred-fold increase in vascular plant richness since the early Creta-ceous (50), as well as major radiations among insects, particularly Hyme-noptera. Mycorrhizae may have been involved in another majorevolutionary event, the invasion of the land by plants (246), and the diver-sity of modern fungi owes a great deal to mycorrhizal and lichenous taxa(4). The existence of coral reefs depends both on endozoic algae (238, 248,294) and on protection mutualisms (102), while the large herbivores grasslands generally depend on ruminant bacteria to maintain their highdensities in these cellulose-rich habitats (202). Deep-sea vent communitiesmay also depend on mutualism (57).

All these arguments for the importance of mutualism in ecosystems aswe know them seem a bit trivial when we turn our attention to the likeli-hood that eukaryotie cells are the descendants of intracellular symbi-onts (205, 206, 207, 326). The serial endosymbiosis theory holds thatmitochondria, chloroplasts, and other organdies are derived fromsymbiotic prokaryotes. Recent discoveries lending credence to this the-ory are reviewed by Whatley et al (326), Margulis (207), and Taylor(295).

Mutualisms are known in all kingdoms of organisms, and there is atendency for the partners to come from different kingdoms (45). This particularly true for obligate and symbiotic mutualisms, and may simply bea reflection of nutritional complementarity. Some taxa seem particularlylikely to enter into mutualisms--e.g. Nostoc, Trebouxia, Symbiodinium,and Chlorella (207), and at a higher level, ants, coelenterates, and legumes.Some taxa participate in several mutualisms simultaneously: Acacia collin-sii, for example, has pollination by bees, dispersal by birds, protection byants, and probably mycorrhizae and Rhizobium nodules (and is also weedy species).

THE NUMBER OF PARTNERS

Three sorts of specificity can be distinguished in examining the interactionof a mutualist with its partners--the numbers of species, of individuals, andof genomes involved. Each of these raises different questions, and all threeare different from the questions of obligacy and symbiosis. However, certaincorrelations seem to emerge.


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



We call the interaction of a single pair of species "monophily." Anexample is the pollination of Yucca glauca by Tegeticula yuccasella (5,251). "Oligophily" is the presence of a few species in each role--say lessthan five. Examples of this interaction are found in, for example, curved-beak hummingbirds and the specialized flowers they pollinate. Finally,"polyphilie" mutualisms involve multiple taxa simultaneously---e.g, yellowcomposites and their pollinators (flies, bees, butterflies).

Many of the most celebrated mutualisms are monophilic: in pollination[Yucca (5, 251), figs and fig wasps (128, 158), solitary bees and their hosts(81, 152, 244)]; ant-plant protection mutualisms [Barteria and Pachysima(154), Cecropia and Azteca (150)]; nutritive mutualisms [Atta and its fungi(255, 337), fungus-gardening beetles (21), at least some lichens (206), vertebrate-alga interactions (207, 222)], and others. Monophily is oftenhard to understand: Two species with clearly separate genomes (jumpinggenes notwithstanding) and independent evolutionary histories interact co-operatively. When monophily is obligate and symbiotic, the two species areoften considered as one. Certainly, the pair ceases to have two sets ofinteractions: Interaction with one species requires interaction with theother. These are the species for which high evolutionary risk has long beenproclaimed. They have given up their "freedom" and depend, evolutionarilyand ecologically, on the presence of a species whose genome evolves inde-pendently. They cannot exchange genes with it, are subject to its mutabilityor lack of mutability, and must endure situations imperiling its survival.Selection will occur, of course, but the mutualist is a victim of the fitnessof its partner, rather than a direct participant.

The conventional wisdom about monophilie obligate mutualists givenabove is an overstatement, of course. Species generally lumber along withnumerous incomplete adaptations to today’s unreliable environment, onlyindifferently adapted to any particular danger. Perhaps mutualistie interde-pendence does not engender major risks very often; whether it raises therisks of extinction over the long run is also unclear. Investigation of thepopulation dynamics of a monophilic mutualism as compared to oligophiliccongeners would be informative.

Monophilic facultative mutualisms are unknown to us. A self-compatibleand self-fertilizing Yucca would constitute one, if such a plant existed. Sincesuch things are by no means impossible, but only unknown, we concludethat they are selected against, perhaps for the reasons mentioned above.There are also mutualisms in which only two species interact at a given sitebut the pairs may vary. For example, Acacia collinsii forms ant-acacias withPseudomyrmex belti, P. ferruginea, and P. nigrophilosa in western CostaRica (149). On an individual level this might be considered monophilic, buton the population level it is certainly oligophilic.


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



Oligophilic mutualists would appear to have the "best of both worlds."Oligophily is the compromise between the risks entailed by specializationand the inefficiency of gcncralist interactions; it is common in both faculta-tire and obligate, symbiotic and nonsymbiotic mutualisms. Much of polli-nation falls into this category, as well as many mycorrhizal associations,nitrogen-fixation, and mixed feeding flocks of birds.

Polyphilic obligate symbiotic mutualism is rare. Invertebrate interactionswith Chlorella may qualify (101, 189, 307); but the taxonomy of symbioticalgae is extremely difficult, so these results may change. Most mutualist taxaare probably polyphilic facultative mutualists, simply because these broadinteractions encompass so many species. Generalist pollination, extrafloralnectades (163, 171), myrmecochory and other animal dispersal of seeds(200, 243), vesicular-arbuscular mychorrhizae (204), and vertebrates their gut flora are examples. Obligate mutualism would appear to operatedifferently from facultative mutualism, with the former tending to be mono-philic or oligophilic, the latter oligophilic or polyphilic. Judging from theabundance of different types, it would seem that facultative mutualisms arereadily established between species but that significant coevolution is rare.The mutualisms remain casual and generalized. In a few cases, coevolutionleads to major fitness gains for the partners, perhaps followed by obligacyand/or symbiosis.

This view is quite different from the one given previously, which sug-gested that symbiotic mutualism evolves from intimate interactions that areparasitic or commensal. It seems likely that both processes have occurred.Perhaps detailed comparison of the (presumed) histories of a variety mutualisms would tell us about the conditions selecting for symbiosis.

In order to take full advantage of the opportunities available once thepartnership is established, the interacting species must have solved thecritical problems of (a) one mutualism and two sets of limiting factors and(b) finding the partner. Often the first problem is solved by the establish-ment of the interaction itself. It is also frequently solved by symbiosis:Living together, both species have the same set of experiences, even if theydo not necessarily react similarly. Thus the symbiotic pair, confronted withthe predator of one, is jointly faced with producing antipredator defensesor perishing. The nonsymbiont, on the other hand (imagine a pollinator),may return to find its partner devoured. The problem of finding the partnercan also be solved by symbiosiS: Putting both mutualists into each propa-gule (whether calf or hydra bud) is an efficient way to maintain the mutual-ism. Consequently, many (but not all) "highly successful" mutualists aresymbiotic.

The numbers of individuals with which one partner is mutualistic rangesfrom one to millions (158). Wilson (336) has discussed this in the light


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



"the classical group selection problem," that of the mutualist’s helping itscompetitors as well as itself. Competition for mutualism has been foundboth within and between species (2, 175, 319), and the within-species com-petition would seem a major obstacle for the evolution of mutualism. Wil-son shows that spatial heterogeneity of interaction (demes structured intotrait-groups) can provide a solution; Post et al’s (249) model of mutualist-commensalist competition gives the same sort of result.

The ratio of numbers of individuals of the two species interacting is animportant consideration, particularly in symbiosis, and it would seem thatthe optimum ratio would not necessarily be the same for both species.Synchronization of reproduction can be seen as a way to maintain a moreor less constant ratio (205, 223), but little is known about how such ratiosvary or the role of conflicts in them.

Symbiotic mutualisms tend to create new "organisms" with two or moredistinct genomes, and the cellular endosymbiosis theory suggests that alleukaryotes are such organisms (207). The dependency involved if the sym-biosis is obligate implies that coordination between genomes, with elimina-tion of redundancy, should be beneficial to the association; but this wouldseem to cordiict with a strict "selfish gene" view. Some examples of coordi-nation are elegant; for example, leghemoglobin in Rhibozium nodules issynthesized partly on the bacterial template (heme) and partly on the planttemplate (globin) (63). Furthermore, the structure of leghemoglobin is markably similar to those of animal hemoglobins. Such cases of mutualismmay provide useful tests for theories of genetic selfishness.

FORMATION AND BREAKDOWN

While some mutualisms are undoubtedly highly coevolved (5, 96, 141, 149,158, 215, 224, 251, 255, 266), and past (and the present) authors haveemphasized evolutionary, processes, mutualisms can easily be formed with-out evolution (157). Species may be preadapted to forming a new mutualismthrough traits presumably evolved with different partners elsewhere; e.g.the extrafloral nectaries of introduced lZicia in California attract the intro-duced ant Iridomyrmex (176), and cattle egrets pick ticks off deer (117).However, other apparent mutualisms have been observed to develop rapidlywith little preadaptation [e.g. dogs and langurs (280), amoebae and bacteria(160)]. Given the facultative, polyphilic nature of many mutualisms, would be an error always to interpret them in coevolutionary terms.

Once a mutualism exists, it is subject to "parasitism" in two senses. First,the association may provide a novel resource on which to prey, such as rootnodules or Acacia thorns. Second, either partner, or a third species, may


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



take one of the benefits normally provided, without giving anything inreturn. Nectar-robbing insects (18, 19, 81, 144), ants preying on their aphids(323), cleaner-fish biting their "hosts" (106), and unaggressive ants inhabit-ing acacias (149) are just a few examples. Acacia-ants may take nectar fromother sources (169), indicating the beginnings of either breakdown of themutualism or oligophily. Invasion of islands, in which one partner may beleft behind, often leads to the loss of coevolved traits (155, 263).

Extinction of one partner ought to have drastic effects on the other ifthere has been substantial coevolution; by the same token, it may result inrapid extinction of the partner and thus leave little evidence behind. In fact,despite the potential importance of this phenomenon (98), few good casesare known (296, but see 234). Loss of a substantial seed disperser commu-nity may simply result in new dispersers’ taking over (159). Neither highlyobligate nor highly facultative mutualisms are likely to give good evidenceof extinction caused by loss of mutualist.

INDIRECT MUTUALISM

Several recent theoretical studies (180, 184, 185, 314, 320, 335, 336)haveindicated that species that never come into physical contact may neverthe-less positively affect each other’s fitnesses or population growth rates. Datasustain these assertions in a substantial number of cases. Nevertheless, mostof the following "indirect mutualisms" should be seen as speculative, inter-esting, and perhaps indicative of the need to revise accepted views of com-munities (261, 336).

Consumer-ResourceUsing MacArthur’s (198) consumer-resource equations, Levine (184) Vandermeer (314) have shown how two consumers, by reducing competi-tion and preventing competitive exclusion among the resources they eat, canbenefit each other. This type of mutualism is a consequence of the interac-tion of two well-accepted ecological phenomena: competition, which canlead to reduced niche overlap (14, 34, 120, 138, 175, 198, 199, 213, 219, 240,254, 312, 329, 330), and the keystone predator effect (60, 65, 99, 121, 122,151, 182, 329). An intriguing aspect of this mutualism is that it followsdirectly from competition and predation. Thus it should be taken seriouslyby ecologists who believe competition and predation to be important instructuring communities.

An obvious consequence of the model of Levine (184) is that indirectmutualism is unlikely to occur in adjacent trophic levels. Plants’ resourcesdo not compete, therefore the plants are not able to engage in this kind of


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



mutualism, and are more likely to be in competition with one another.Herbivores could be indirect mutualists, but this would mean that theanimals which feed on those herbivores would not be.

A promising system for the verification of a case of indirect mutualismexists in subalpine ponds in Colorado (72). The distribution of a midge larva(Chaoborus) in these ponds has been shown to depend on the presence ofa larval salamander (Ambystoma), apparently through interactions betweenthe large and small species of Daphnia, which are eaten by Ambystoma andChaoborus, respectively. Large Daphnia were present in small numbers andsmall Daphnia in large numbers in ponds where the salamander lived. Sinceabsence of salamanders is correlated with absence of small Daphnia, midgescan only survive in ponds inhabitated by salamanders. Up to this point, thesystem appears to fit the requirements for indirect mutualism outlinedabove: The two consumers have different feeding habits and the resourcespecies probably compete for food (both Daphnia are herbivores). To dem-onstrate mutualism it remains to manipulate population densities of all fourspecies in controlled conditions.

Possible consumer-resource indirect mutualisms also exist among terres-trial herbivores. Two herbivorous rodents studied by Cameron (54) weresubjected to removal experiments. Results indicated a mutualistic relation-ship, as survivorship and reproduction were lower in both experimentalpopulations than in control populations, but the mechanism was not deter-mined. On the Serengeti plain in Africa, where wildebeest, Thompson’sgazelle, and zebra are the large herbivores, there is another possible case ofindirect mutualism. Stomach contents indicate a dietary separation amongthe three (114). Thompson’s gazelles prefer to feed in areas where wilde-beest have grazed one month previously, since these regions have greaterplant biomass (202). No data on the effects of gazelle grazing on the quan-tity of grasses to be eaten by wildebeest are available. This appears to bea promising case, but the full story is not yet told.

Enemies’ Enemies

Indirect mutualism may also occur among competing species, as shown byLawlor (180). If species A competes with B, and B with C, the net interac-tion between A and C may be mutualistic, and the principle can be general-ized to multispecies communities. Lawlor’s (180) analysis of niche overlapmatrixes for bird communities indicates such mutualism, and Seifert &Seifert (278) found mutualisms in Heliconia bract invertebrate communitiesthrough removal experiments. The principle of "two negatives may createa positive" can be extended to communities containing any kind of interac-tion, using Levins’s loop analysis techniques (185). Here again, species with


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



no direct interaction may affect each other’s growth rates positively. Themethod is particularly powerful in that it generates testable predictionsusing only knowledge of the signs of interactions.

Friends’ Friends

The reverse of the above situation can also exist: if A and C are bothmutualistic with B, A and C can benefit each other indirectly. This possibil-ity, which also can be derived from loop analysis, is indicated in a numberof pollination systems. Mimicry among flowers in limited populations, suchas those found in alpine areas, could enlarge the population size perceivedby pollinators. This has been proposed by Macior (203) for a number sympatric, synchronously blooming species that resemble one another andshare a pollinator. Several subalpine species in Colorado, all of which arepollinated by butterflies, share an ultraviolet reflectance pattern (322).Brown & Kodric-Brown (49) have described a system in which seven spe-cies with flowers similar in appearance place pollen on different parts of thepollinators, in this case humingbirds. If mimicry did not occur, constantpollinators would switch from rare species to rare species upon perceivingthat each is insufficiently rewarding, until they landed on a common specieswith which they would stay, to the detriment of the rare species (37). Somepollen is wasted when pollinators move between mimics, but this may favora plant more than not being visited at all.

While competition for pollinators is thought to explain divergent flower-ing times (219, 319), the consequences of divergence may be a form indirect mutualism (16, 18, 272). This has been demonstrated by Waser Real (320) on the same species studied in the competition experiments Waser (319). When the number of flowers on the early-flowering species wasreduced, the seed set of the later-flowering species declined. In the short run,this works in one direction, to the benefit of the later-flowering species. Inthe long view, both species support the pollinator population, and it isreasonable to hypothesize that removing either species will reduce futurepollination by reducing the resources available to the pollinators. While theabove example is from the temperate zone, the phenomenon may be wide-spread; Schemske (272) describes "pollinator sharing" in tropical herbs.The general principle is that the increase in pollination must outweigh theloss of pollen in interspecific transfers. Mutualist maintenance over longerperiods or in higher densities than would occur if fewer mutualistic specieswere present may indeed apply to many other systems (18, 147).

This raises another question: When does sharing a mutualist result incompetition and when does it result in indirect mutualism among the


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



sharers? Either is possible; and at present we have little knowledge, eithertheoretical or empirical, with which to answer.

Not all the interactions in a "friends’ friend" mutualism need be positive.Somewhat more complicated cases are found when ant protection of aphidsalso results in protection of the plants on which they feed (214). Theimportant point is that the ants protect the plant and the plant indirectly(via the aphid) feeds the ants. One could extend this to systems in whichthe ants simply prey on insects found on the plant and thus protect it,without any direct mutualism between ant and herbivore at all. This resultsin "green islands" of protected plants in the vicinity of wood ant nests (179).Indeed Price et al (251a) suggest that the role of plant defenses is oftenindirect, through making herbivores more vulnerable to predation andparasitism; thus "Enemies should be considered as mutualists with plants."

Protection without Interaction

A final class of indirect mutualism involves species whose interactions withthird species tend to reduce predation on each other. The third species maybe a predator whose rate of predation on each of two prey is reduced whenboth are present. This may be due to such phenomena as Miillerian mimicryor predator satiation (153). It may also result from chemical protection,such as in the "plant defense guilds" proposed by Atsatt & O’Dowd (10).In all these cases, a simple safety-in-numbers (whatever the species to whichthe "numbers" belong) is all that is necessary to produce indirect mutualismby reducing the probability that a given prey individual is eaten.

A more complicated situation is described by Lubchenco (196) in rockyintertidal communities. A green crab eats young periwinkles, which fedupon the alga Enteromorpha. Enteromorpha provides good cover for thecrabs, hiding them from gulls. Where periwinkles are not held in check, thealga Chondrus becomes dominant, but crabs cannot hide well in this alga.There is no known direct interaction of crab and Enteromorpha, so this isan indirect mutualism.

While we have no reason to believe that new evolutionary theories willbe necessary to explain the evolution of any kind of indirect mutualism, itis appropriate to point out the work of Wilson in this connection (335, 336).Wilson’s model depends on the existence of spatial variation within commu-nities and thus differential feedback through loops of interactions. Thiscondition satisfied, two genotypes may be differently affected by their in-teractions with other elements of the community. Thus species or genotypescan be selected to increase the abundance of other species or genotypes ifthis brings positive feedback. Clearly, this could result in the evolution ofindirect mutualisms, and in simulations it produces many mutualistic in-


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



teractions (336). The question that remains is whether Wilson’s modelaccounts for anything existing in the real world. This brings us back to theneed for more information on the abundance of mutualistic interactions,particularly those that appear to be coincidental or the outcome of evolutiondriven by other processes.

FUTURE DIRECTIONS

Different approaches to ecology have emphasized different interactions andlevels of organization. Ecosystem studies tend to look at the interconnected-hess of the species in a community with respect to their roles in energy flowand nutrient cycles. This emphasizes wholeness and beneficial relationships,or at least takes a positive view of negative interactions as part of ecosystemfunctioning. The other major school of thought in ecology is much more-individualistic in approach and has risen to preeminence with the declineof the "community as superorganism" view. After a period of being hauledout of the past and held up as a bad example of scientific practice, theorganismic school has been given a tentative pardon in a commentary byRichardson (261). Richardson proposes that the organismic view is sup-ported by evidence of the existence of "multispecies group mutualisms," orof species that play a key role in giving a community its distinctness. Itremains to be seen whether future work on mutualism will help to reconcilethe organismic and individualistic views. When some workers see mutual-ism as an integrating mechanism for whole communities (261, 336), andothers view it as the endpoint of a "mutually exploitative arms race" (67),there is obviously a philosophical gap of some size to be bridged.

What is clear is that the study of mutualism has made major advancesin just the past decade. Theorists have successfully defended mutualismagainst the charges of being destabilizing (211, 212) and group-selectionist(333), while field studies have shown it to be widespread and important many population and community characteristics (17, 33, 85, 96, 108, 109,115, 152, 244, 258, 287, 300, 327, 337, 338). Given the importance of manymutualisms to human welfare (40, 51, 87, 98, 134, 136, 204, 210, 256, 339),not to mention their elegance and beauty, we can only hope that the rapidgrowth of interest in mutualism will continue.

ACKNOWLEDGMENTS

Our thanks to A. Cambrosio, J. Caron, R. Heithaus, D. Janzen, M.Lechowicz, S. Levin, S. Risch, D. Simberlotf, J. Vandermeer, D. Wilson,and especially E. Fung-A-Ling and S. Marehand, for their help.


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



Literature C~’ted

1. Addicott, J. F. 1978. Competition formutualists: aphids and ants. Can. J.Zoo/. 56:2093-96

2. Addicott, J. F. 1979. A multispeciesaphid-ant association: density depen-dence and species-specific effects. Can.J. Zoo/. 57:558-59

3. Addicott, J. F. 1981. Stability proper-ties of two-species models of mutualism:simulation studies. Oecologia 49:42-49

4. Ahmadjian, V. 1967. The Lichen Sym-biosi~ Waltham, MA: Blaisdell

5. Aker, C. L., Udovic, D. 1981. Oviposi-tion and pollination behavior of theyucca moth, Tegeticula maculata

fi(Lepidoptera: Prudoxidae), and its rela-on to the reproductive biology ofYucca whipplei (Agavaceae). Oecologia49:96-101

6. Allee, W. C., Emerson, A. E., Park, O.,$¢hmidt, K. P. 1949. Principles of Ani-malEcology. Philadelphia: W. B. Saun-ders

7. Allen, E. K., Allen, O. N. 1959. Thescope of nodulation among members ofthe family Lcguminosae. Pro~ Intl. Bot.Congr., 9th 2:3-4

8. Armstrong, R. A. 1978. The coloniza-tion of a slate rock surface by a lichen.New PhytoL 81:85-88

9. Atsatt, P. R. 1981. Lycaenid butterfliesand ants: selection for enemy-free space.,4ra. Nat. 118:634-54

10. Atsatt, P. R., O’Dowd, D. J. 1976.Plant defense guilds. Science 193:24-29

11. Altmann, S. A., Altmann, J. 1970. Ba-boon Ecology: African ~’)’eld Researc&NY: S. Kin-get

12. Auelair, J. L. 1963. Aphid feeding andnutrition. Ann. Rev. EntomoL 8:439-90

13. Axelrod, R., Hamilton, W. D. 1981.The evolution of cooperation. Science211:1390-96

14. Ayala, F. J., Gilpin, M. E., Ehrenfeld,J. G. 1973. Competition between spe-cies: theoretical models and experimen-tal tests. Theor. Pop. BioL 4:331-56

15. Bach, C. E., Herrnkind, H. F. 1980.Effects of predation pressure on the mu-tualistic interaction between the hermitcrab Pagurus pollicaris Say, 1817, andthe sea anemone Calliactus tricolor (LeBueur, 1817). Crustaceans 38:104-8

16. Baker, H. G. 1963. Evolutionary mech-anisms in pollination biology. Science139:877-83

17. Baker, H. G. 1973. Evolutionary rela-tionships between flowering plants andanimals in American and African tropi-cal forests. In Tropical Forest Ecosys-tems in Africa and South America, ed.

B. J. Mcggers, E. S. Ayensu, W. D.Duckworth. Washington DC: Smith-sonian Inst. Press

18. Baker, H. O., Cruden, R. W., Baker, I.1971. Minor parasitism and its commu-nity function: the case of Ceibaaeuminata. BioScienee 21:1127-29

19. Baker, J. M. 1963. Ambrosia beetlesand fungi. In Symbiotic Associations,ed. P. S. Nutman, B. Mosse.Cambridge: Cambridge Univ. Press

20. Balda, R. P. 1980. Are seed caching sys-tems co-evolved? Act. Congr. Int. Orni-thoL, 17th, pp. 1185-91

21. Batra, L. R., ed. 1979. Insect-FungusSymbiosis: Nutrition, Mutualism, andCommensalis~ NY: Wiley

22. Beattie, A. J., Culver, D. C. 1981. Theimportance of myrmeeochory in theherbaceous flora of West Virginia for-ests. Ecology 62:107-15

23. Becking, J. H. 1979. Root nodule sym-biosis between Rhizobium and Para-sponia (Ulmaceae). Plant Soil 51:289-96

24. Belt, T. 1874. The Naturalist in Nicura-gua. London: J. Murray. 403 pp.

25. Benemann, J. R. 1973. Nitrogen fixa-tion in termites. Science 181:164-65

26. Bentley, B. L. 1976. Plants bearing ex-trattoral hectares and the associated antcommunity: interhabitat differences inthe reduction of herbivore damage.Ecology 57:815-20

27. Bentley, B. L. 1977. Extralloral nectar-ies and protection by pugnacious body-guards. Ann. Rev. Ecol. Syst. 8:407-27

28. Bentley, B. L. 1977. The protectivefunction of ants visiting the extrattoralnectaries of Bixa orellana L. O~ix-aceae). Z Ecol. 65:27-38

29. Bequaert, J. 1922. Ants in their diverserelation to the plant world. Bull Am.Mu£ Na~ Hist. 45:333-583

30. Berg, R. Y. 1975. Myrmecochorousplants in Australia and their dispersalby ants. Austral .L Bot. 68:478-502

31. Beringer, J. E., Brewin, N., Johnston,A. W. B., Schulman, H. M., Hopwood,D. A. 1979. The Rhizobium-legumesymbiosis. Proc. R. Soc. Lond. Set. B204:219-33

32. Bernstein, I. S. 1967. Intertaxa interac-tions in a Malayan primate enmmunity.Folia PrimatoL 7:198-207

33. Bierzychudek, P. 1981. Pollinator limi-tation of plant reproductive effort. Arr~Nat. 117:838-40

34. Black, J. N. 1960. An assessment of therole of planting density in competitionbetween Red Clover (Trifolium pratense


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



L.) and Lucerne (Medicago sativa L.) inthe early vegetative stands. Oikos11:2~42

35. Blom, P. E., Clark, W. H. 1980. Obser-vations of ants fflymenoptera: For-micidae) visiting extrafloral nectaries ofthe barrel cactus Ferrocactus gracilis(Caetaceae) in Baja California, Mexico.Southw. Nat. 25:181-96

36. Bloom, S. A. 1975. The motile escaperesponse of a sessile prey: sponge scal-lop mutualism. J. Exp. Mar. Biol.17:311-28

37. Bobisud, L. E., Neuhaus, R. J. 1975.Pollinator constancy and survival ofrare species. Oecologia 21:263-72

38. Bond, G. 1963. The root nodules ofnon-leguminous angiosperms. In Sym-biotic Associations, ed. P. S. Nutman, B.Mosse. Cambridge: Carabddge Univ.Press

39. Boucher, D. H. 1979. Seed predationand dispersal by animals in a tropicaldry forest. PhD thesis. Univ. Michigan,Ann Arbor

40. Boucher, D. H. 1979. La nodulacion ddfrijol en el policultivo I. E1 efecto de ladistancia entre plantas de frijol y maiz..4gric. Trop. (CSAT) 1:276-83

41. Boulding, K. E. 1972. Economics as anot very biological science. In Challeng-ing Biological Problems, ed. J. A.Behnke, pp. 357-75. Oxford: OxfordUniv. Press

42. Bowen, G. D. 1978. Dysfunction andshortfalls in symbiotic responses. InPlant Disease, ed. J. G. Harsfall, E. B.Cowling, Vol. 3. NY: Academic

43. Boyle, C. D., Patdquin, D. G. 1980.Eodorhizal and exorhizal acetylene-reducing activity in a grass (Spartinaalternifloro Loisd.)-diazotroph associa-tion. Plant Physiol. 66:26-280

44. Breznak, L A., Brin, W. L, Mertins, J.W., Coppd, H. C. 1973. Nitrogen fixa-tion in termites. Nature 244:577-79

45. Briand, F., Yodzis, P. 1982. The phylo-genetic distribution of obligate mutual-ism. Oiko~ In press

46. Bristow, C. M. 1982. The structure of atemperate zone mutualism: ants and lto-moptera on New York Ironweed. PhDthesis. Princeton Univ.

47. Brock, T. D. 1966. Principles of Mi-crobial Ecology. Englewood Cliffs, NJ:Prentice-Hall

48. Brooks, M. A. 1963. Symbiosis and apo-symbiosis in arthropods. In SymbioticAssociations, ed. P. S. Nutman, B.Mosse, pp. 200-31. Cambridge: Cam-bridge Univ. Press

49. Brown, J. H., Kodrie-Brown, A. 1979.Convergence, competition, and mim-icry in a temperate community of hum-mingbird-pollinated flowers. Ecology60:1022-35

50. Burger, W. C. 1981. Why are there somany kinds of flowering plants? Bio-Science 31:572, 577-81

51. Buchner, P. 1965. Endosymbiosis of An-imals with Plant Microorganisms. NY:Wiley

52. Buss, L. W. 1981. Group living, compe,tition and the evolution of cooperationin a sessile invertebrate. Science213:1012-14

53. Callanham, D., Del Tredici, P., Torrey,J. G. 1978. Isolation and cultivation invitro of the actinomycete causing rootnodulation in Comptonict Science199:899-902

54. Cameron, G. N. 1975. Interspeeifie in-teractions: a case of facilitation. Atystr.Ann. Meet. ,qtr~ Soc. Mammalogists,55th: 26

55. Carpenter, E. J., Culliney, J. L. 1975.Nitrogen fixation in marine shipworms.Science 187:551-52

56. Cates, R. G., Orians, G. H. 1975. Sue-cessional status and the palatibility ofplants to generalized herbivores.Ecology 56:410-18

57. Cavanaugh, C. M., Gardener, S. L.,Jones, M. L., Jannasch, H. W., Water-bury, J. B. 1981. Prokaryotic cells in thehydrothermal vent tube worm Rifliapachyptila Jones: possible ehemoauto-trophic symbionts. Science 213:340-42

58. Clark, K. L., Robertson, R. J. 1979.Spatial and temporal multi-species nest-ing aggregations in birds as anti-para-site and anti-predator defenses. Behav.Ecol. SociobioL 5:359-71

59. Cody, M. L. 1971. Finch flocks in theMojave Desert. Theor. Pop. Biol. 2:142-58

60. Connell, J. H. 1975. Some mechanismsproducing structure in natural commu-nities: a model and evidence from fieldexperiments. In Ecology and Evolutionof Communities, ed. M. L. Cody, J. M.Diamond. Cambridge, MA: BelknapPress

61. Connor, R. C., Norris, K. S. 1982. Aredolphins reciprocal altruists? Am. Nat.119:358-74

62. Cooke, R. 1977. The Biology of Syrabi-otic Fung£ NY: Wiley. 282 pp.

63. Cutting, J. A., Schulman, H. M. 1972.The control of heme synthesis in soy-bean root nodules. Biochi~ BiophyxActa 229:58-62

64. Deleted in proof


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



65. Darnell, R. M. 1970. Evolution and theecosystem. Am. Zool. 10:9-15

66. Davidson, D. W., Morton, S. R. 1981.Myrmecochory in some plants (f.Chenopodiaceae) of the Australian aridzone. Oecologia 50:357-66

67. Dawkins, R., Krebs, J. R. 1979. Armsraces between and within specie*. Proc.R. Soc. Lond. Ser. B 205:489-511

68. DeAngelis, D. L., Travis, C. C., Post.,W. M. III. 1979. Persistence and stabil-ity of seed-dispersed species in a patchyenvironment. Theor. Pop. Biol. 16:.107-25

69. DeBary, A. 1879. Die Erscheinung derSymbiose. Strasburg: Karl J. Trubner

70. Delpino, F. 1886-1889. Funzione mir-mecoflle nel regno vegetale. Mem. R.Accad. Sic. Bologna, Ser. 4. 7:215-392;8:602-59; 10:155-247

71. Dilworth, M. J., Parker, C. A. 1969.Development of the nitrogen-fixing sys-tem in legumes. J. Theor. Biol.25:208-18

72. Dodson, S. I. 1970. Complementaryfeeding niches sustained by size selec-tive predation. Limnol. Oceanogr. 15:131-37

73. Dubos, R., Kessler, A. 1963. Integra-tive and disintegrative factors in symbi-otic associations. In Symbiotic Associa-tions, ¢d. P. S. Nutman, B. Musse.Cambridge: Cambridge Univ. Press

74. Duggins, D. O. 1981. Interspccific facil-itation in a guild of benthic marine her-bivores. Oecologia 48:157-63

75. Dyer, M. I., Bokhari, V. G. 1976. Plantanimal interactions: studies of theeffects of grasshopper grazing on bluegrama grass. Ecology 57:762-72

76. Egerton, F. N. 1977. History of~meri-can Ecology. NY: Arno Press. 271 pp.

77. Ehrlich, P. P., Ehrlieh, A. H. 1973. Co-evolution: heterotypic schooling inCaribbean reef fishes. Ara. Nat.107:157-60

78. Elder, W. H., Elder, N. L. 1970. Social

t~rloupings and primate associations ofe bushbuck (Tragelaphus scriptus).Mammalia 34:356-62

79. Epp, R. W., Lewis, W. M. Jr. 1981.Photosynthesis in eopepods. Science214:1349-50

80. Espinas, A. V. 1878. Des Societes ~ni-males~ Paris: Bailliere

81. Faegri, K., van der Pijl, L. 1966. Princi-pies of Pollination Ecology. Oxford: Per-gamon Press

82. Fahn, A. 1979. Secretory Tissues inPlant~ NY: Academic

83. Feder, H. M. 1966. Cleaning symbiosisin the marine environment. In Symbio-

sis, ed. S. M. Henry, Vol. 1. NY; Aca-demic

84. Fenchel, T., Christiansen, F. B. 1976.Theories of Biological Communities.NY: Springer

85. Fleming, T. H., Heithaus, E. R. 1981.Frngivorons bats, seed shadows, andthe structure of tropical forests. Bio-tropica 13(2):Suppl. 45-53

86. Florkin, M. 1949. Biochemical Evolu-tion. NY: Academic

87. Fogel, R. 1980. Mycorrhizae and nutri-ent cycling in natural forest ecosystems.New Phytol. 86:199-212

88. Fogel, R., Trappe, J. M. 1979. Fungusconsumption (mycophagy) by small an-imals. Northw. ScL 52:1-31

89. Fricke, H. W. 1975. The role of behav-ior in marine symbiotic animals. Symp.Soc. Exp. Biol. 29:581-94

90. Friedmarm, H. 1967. Avian symbiosis.In Symbiosis, ed. S. M. Henry, Vol. 2.NY: Academic

91. Futuyma, D. J. 1973. Communitystructure in a constant environment.Arr~ NaL 107:443-46

92. Garling, L. 1979. Origin of ant-fungusmutualism: a new hypothesis. Bio-tropica 11:284-91

93. Gause, G. F. 1934. The Struggle for Ex-istence. Baltimore: Williams and Wat-kins

94. Gause, G. F., Witt, A. A. 1935. Behav-ior of mixed populations and the prob-lem of natural selection. Ar~ NaL69:596-609

95. Gilbert, D. G. 1980. Dispersal of yeastsand bacteria by Drosophila in a temper-ate forest. Oecologia 46:135-37

96. Gilbert, L. E. 1975. Ecological conse-quences of a co-evolved mutualism be-tween butterflies and plants. In Coevolu-tion of Animals and Plants, ed. L. E.Gilbert, P. H. Raven. Austin, TX: Univ.Texas Press

97. Gilbert, L. E. 1979. Development oftheory in the analysis of insect-plant in-teractions. In Analysis of Ecological Sys-tems, ed. D. Horn, R. Mitchell, G.Stairs, pp. 117-54. Columbus: OhioState Univ. Press

98. Gilbert, L. E. 1980. Food web organiza-tion and the conservation of neotropicaldiversity. In Conservation Biology, ed.M. E. Soule, B. A. Wilcox, Ch. 2. Sun-derland, MA: Sinauer Associates

99. Gilpin, M. E. 1972. Enriched predator-prey systems: theoretical stability.Science 177:902-4

100. Gilpin, M. E., Case, T. J., Bender, E. A.1982. Counterintuitive oscillations in


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



systems of competition and mutualism.Am- Nat. 119:584-88

101.Glider, W. V., Phipps, D. W., Pardy, R.L. 1980. Localization of symbioticdinoflagellate cells with tentacle tissueof Aiptasia pallida (Coelenterata, An-thozoa). Trans. din. Micros. Soc.90:426-38

102.Glynn, P. W. 1976. Somephysical andbiological determinants of coral com-munity structure in the eastern Pacific.Ecol. Monogr. 46:431-56

103.Goh, B. S. 1979. Stability in models ofmutualism. Am. Nat. 113:261-75

104.Goldman, P. 1980. Flocking as a possi-ble predator defense in dark-eyed jun-cos. Wilson Bull. 92:88-95

105.Goulding, M. 1980. The Fishes and theForest. Berkeley: Univ. Calif. Press.250 .pp.

106.Gorhck, D. L., Atkins, P. D., Losey, G.S. 1978. Cleaning stations as waterholes, garbage dumps, and sites for theevolution of reciprocal altruism. An~Nat. 112:341-53

107.Granhail, U., yon Hofsten, A. 1976.Thenitrogenase activity in relation to in-tracellular organisms in Sphagnummosses. Physiol. Plant. 36:88-94

108.Grant, V. 1949. Pollination systems asisolating mechanisms in Angiosperms.Evolution 3:82-97

109.Grant, V., Grant, K. A, 1965. FlowerPollination in the Phlox Family. NY:Columbia Univ. Press

110.Grinfel’d, E. K. 1977. Anthophily inbeetles (Coleoptera) and a critical eval-uation of the eantharophilous hypothe-sis. Entomol. Rev. 54(3):18-22

111.Grossman, G. D. 1979. Symbiotic bur-row-occupying behavior in the bay gobyLepidogobius lepidus. Calif. Fish. 65:122-24

112.Guerinot, M. L., Fong, W., Patriquin,D. G. 1977. Nitrogen fixation (acetylenereduction) associated with sea urchinsStrongylocentrotus droebachensis) feed-ing on sea weeds and eelgrass. J. Fish.Rex Bd. Cam 34:416-20

113.Gutschick, V. P. 1981. Evolved strate-gies in nitrogen acquisition by plants.Am. Nat. 118:607-37

114.Gwynne, M. D., Bell, R. H. V. 1968.Selection of vegetation components bygrazing ungulates in the Serengeti Na-tional Park. Nature 220:390-93

115.Hacskaylo, E., Tompkins, C. M. 1973.World literature on mycorrhizae. Con-trib. Reed Herb. 22:

116.Hallam, T. G. 1980. Effects of cooper-ation on competitive systems. J. Theor.BioL 82:415-23

117.Halley, M. R., Lord, W. D. 1978. Acattle egret-deer mutualism. WilsonBull. 90:291

118.Hamilton, W. D. 1971. Geometry forthe selfish herd. Z Theor. Biol. 31:295-311

119.Handel, S. N. 1978. New ant-dispersedspecies in the genera Carex, Luzula,and Claytonia. Can. J. Bot. 56:2925-27

120.Harper, J. L. 1963. The nature and con-sequence of interference, amongstplants. Proc. lntl. Congr. Genet., 11th2:465-81

121.Harper, J. L. 1969. The role of preda-tion in vegetational diversity. In Diver-sity and Stability in Ecological Systems(Brookhaven Syrup. tiiol. 22), ed. G. M.Woodwell, H. H. Smith. Springfield,VA: US Sci. Tech. Info. Serv.

122.Harper, J. L. 1977. Population Biologyof Plants. NY: Academic

123.Hartzell, A. 1967. Insect endosymbio-sis. In Symbiosis, ed. S. M. Henry, Vol.2. NY: Academic

124.Heithaus, E. R. 1981. Seed predation byrodents on three ant-dispersed plants.Ecology 62:136-45

125.Heithaus, E. R., Culver, D. C., Beattie,A. J. 1980. Models of some ant-plantmutualisms. Am. Nat. 116:347-61

126.Herring, P. J. 1978. l~ioluminescence in.4ction. N-Y: Academic

127.Hickman, J. C. 1974. Pollination byants: a low-energy system. Science184:1290-92

128.Hilbert, D. W., Swift, D. M., Detling, J.K., Dyer, M. I. 1981. Relative growthrates and the grazing optimization hy-pothesis. Oecologia 51:14-18

129.Hill, D. S. 1967. Fi~s (Ficusspp.) andfig-wasps (Chalcidmdea). J.. Nat. Hist.1:413-34

130.Hinton, H. E. 1951. MyrmecophilousLycaenidae and other Lepidoptera--asummary. Tran~ S. Lond. Entomol.Nat. Hist, Soc. 1949-1950:111-75

131.Hobson, E. S. 1969. Comments on cer-tain recent generalizations regardingcleaning symbiosis in fishes. Pac. ScL23:35-39

132.Hocking, B. 1975. Ant-plant mutu-alism: evolution and energy. In Coevolu-tion of Animals and Plants, ed. L. E.Gilbert, P. H. Raven, pp. 78-90. Ans-tin, TX: Univ. Texas Press

133.Horvitz, C. C., Beattie, A. J. 1980. Antdispersal of Calathea (Marantaeeae)seeds by carnivorous Ponerines (For-micidae) in a tropical rain forest. Am. 1lot. 67:321-26

134.Howard, B. H. 1967. Intestinal micro-organisms of ruminants and other verte-


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



brates. In Symbiosis, ed. S. M. Henry,Vol. 2. NY: Academic

135. Howe, H. F. 1977. Bird activity andseed dispersal of a tropical wet foresttree. Ecology 58:539-50

136. Hungate, R. E. 1966. The Rumen andits Microbes. NY: Academic

137. Deleted in proof138. Hutchinson, G. E. 1957. Concluding re-

marks. CoM Spring Harbor Symp.Quant. Biol. 22:415-27

139. Hutchinson, G. E. 1978. An Introduc-tion to Population Ecology. New Haven,CT: Yale Univ. Press

140. Huxley, C. 1978. The ant-plantsMyrmecodia and Hydnophytum(Rubiaceae) and the relationships be-tween their morphology, ant occupants,physiology and ecology. New Phytol.80:231-68

141. Huxley, C. 1980. Symbiosis betweenants and epiphytes. BiolRe~. 55:321-40

142. Deleted in proof143. Inouye, D. W., Taylor, O. R. 1979. A

temperate region plant-ant-seed preda-tor system: consequences of extratioralnectar secretion by Helianthella quin-quenerviz Ecology 60:1-7

144. Inouye, D. W. 1980. The terminology offloral larceny. Ecology 61:1251-53

145. Institute for Scientific Information.1965-1979. Science Citation lndex.Philadelphia: Inst. Sci. Inf.

146. Janos, D. P. 1975. Effects of vesicular-arbuscular mycorrhizae on lowlandtropical rainforest trees. In En-domycorrhizas, ed. F. E. Sanders, B.Mosse, P. B. Tinker, pp. 437-46. NY:Academic

147. Janos, D. P. 1980. Vesicular-arbuscularmycorrhizae affect lowland tropicalrain forest plant growth. Ecology61:151-62

148. Janson, C. H., Terborgh, J., Emmons,L. H. 1981. Non-flying mammals arepollinating agents in the Amazonianforest. Biotropica 13(2): Suppl. 1-6

149. Janzen, D. H. 1966. Coevolution ofmu-tualism between ants and acacias inCentral America. Evolution 20:249-75

150. Janzen, D. H. 1969. Allelopathy bymyrmecophytes: the ant -4zteca as anallelopathic agent of Cecropio~ Ecology50:147-52

151. Janzen, D. H. 1970. Herbivores and thenumber of tree species in the tropics.Am. Nat. 104:501-28

152. Janzen, D. H. 1971. Euglossine bees aslong-distance pollinators of tropicalplants. Science 171:203-5

153. Janzen, D. H. 1971. Seed predation byanimals, dnr~ Rev. Ecol. Syst. 2:465-92

154. Janzen, D. H. 1972. Protection of Bar-teria (Passitioraceae)by Pachysimaants in a Nigerian rain forest. Ecology53:885-92

155. Janzen, D. H. 1973. Dissolution ofmu-tualism between Cecropia and its Aztecaants. Biotropica 5:15-28

156. Janzen, D. H. 1974. Epiphytie myr-mecophytes in Sarawak: mutualismthrough the feeding of plants by ants.Biotropica 6:237-59

157. Janzen, D. H. 1980. When is it coevolu-tion? Evolution 34:611-12

158. Janzen, D. H. 1979. How to be a fig.Ann. Rev. Ecol. Syst. 10:13-51

159. Janzen, D. H., Martin, P. S. 1982. Neo-tropical anachronisms: the fruits thegomphotheres ate. Science 215:19-26

160. Jeon, K. W., Jeon, M. S. 1976. En-dosymbiosis in amoebae: recently estab-lished endosymbionts have become re-

~hysUiredcytoplasmic components. J. Celliol. 89:337~14

161. Joshi, M. M., Hollis, J. P. 1977. Interac-tion of Beggiatoa and rice plant: detox-ification of hydrogen sulfide in the ricerhizosphere. Science 195:179-80

162. Kalin Arroyo, M. T. 1979. Commentson breeding systems in neotropical for-ests. In Tropical Botany, ed. K. Larsen,L. B. Holm-Nielsen, pp. 371-80. NY:Academic

163. Keeler, K. H. 1977. The extra.floral nec-taxies of lpomea carnea (Convol-vulaceae). Am. J. Bot. 64:1182-88

164. Keeler, K. H. 1979. Distribution ofplants with extrafloral nectaries andants at two elevations in Jamaica. Bio-tropica 11:152-54

165. Keeler, K. H. 1979. Species with extra-floral nectaries in a temperate flora (Ne-braska). Prairie Nat. 11:33-38

166. Keeler, K. H. 1980. The distribution ofplants with extrafloral nectades andants in temperate ecosystems. //m.Midl. Nat. 104:274-80

167. Keeler, K. H. 1981. Cover of plantswith extrafloral nectaries at four North-eru California sites. Madrono 28:26-29

168. Keeler, K. H. 1981. Function of Ment-zelia nuda (Loasaceae) postfloral nec-taries in seed defense. Am. J. Bot.68:295-99

169. Keeler, K. H. 1981. Infidelity by acacia-ants. Biotropica 13:79-80

170. Keeler, K. H. 1981. A model of sdec-tion for facultative, non-symbiotic mu-tualism. Arr~ Nat. 118:488-98

171. Keeler, K. H., Kaul, R. B. 1979. Themorphology and distribution of petiolarnectaries in lpomoea (Convolvulaceae).Arn~ J. Bot. 66:946-52


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



172. Kemna, A. 1897. P. J. van Beneden, lavie et l’oeuvre d’un zoologiste. Anvers: J.E. Buschmann. 137 pp.

173. Kleinfeldt, S. E. 1978. Ant-gardens: theinteraction of Codonanthe crassifolia(Gesneriaceae) and Crematogaster lon-gispina (Formicidae). Ecology 59:449-56

174. Koch, A. 1967. Insects and their en-dosymbionts. In Symbiosis, ed. S. MHenry, Vol. 2. NY: Academic

175. Kodrie-Brown, A., Brown, J. H. 1979.Competition between distantly relatedtaxa in the coevolution of plants andpollinators. Am. Zool. 19:1115-27

176. Koptur, S. 1979. Facultative mutualismbetween weedy vetches bearing extra-floral nectaries and weedy ants in Cali-fornia. Am. J. Bot. 66:1016-20

177. Kostitzin, V. A. 1934. Symbiose, Parasi-tisme, et Evolution (Etude Math-ematique). Paris: Hermann et Cie.

178. Kropotkin, P. 1902. Mutual Aid." A Fac-tor in Evolutior~ London: Heinemann

179. Laine, K. J., Niemela, P. 1980. The in-fluence of ants on the survival of moun-tain birches during an Oporina autum-nata (Lep., Geometddae) outbreak.Oecologia 47:39-42

180. Lawlor, L. R. 1979. Direct and indirecteffects of n-species competition.Oecologia 43:355-64

181. Lazarus, J. 1979. Flock size and behav-iour in captive red-billed weaver birds(Quelea quelea): implications for socialfacilitation and the functions of flock-ing. Behaviour 71:127-45

182. Levin, S. A., Paine, R. A. 1974. Distur-bance, patch formation and communitystructure. Proc. Natl. Acad. Sci. USA71:2744-47

183. Levin, S. A., Udovic, J. D. 1977. Amathematical model of coevolving pop-ulations. Am. Nat. 111:657-75

184. Levine, S. H. 1976. Competitive in-teractions in ecosystems. Am. Nat.110:903-10

185. Levins, R. 1975. Evolution in commu-nities near equilibrium. In Ecology andEvolution of Communities, ed. M. L.Cody, J. M. Diamond. Cambridge, MA:Belknap Press

186. Deleted in proof187. Lewis, D. H. 1973. The relevance of

symbiosis to taxonomy and ecology,with particular reference to mutualisticsymbioses and exploitation of marginalhabitats. In Taxonomy and Ecology, ed.V. H. Heywood. NY: Academic

188. Limbaugh, C. 1961. Cleaning symbio-sis. Scl. Am. 205:42-49

189. Loeblich, A. R. III, Sherley, J. L. 1979.Observations on the theca of the motilephase of free-living and symbiotic iso-lates of Zooxanthella amicroadriatica(Freudenthal) comb. nov. J. Mar. Biol.Assoc. U. K. 59:195-205

190. Losey, G. S. 1972. The ecological im-portance of cleaning symbiosis. Copeia1972:820-23

191. Losey, G. S. 1974. Cleaning symbiosisin Puerto Rico with comparison to thetropical Pacific. Copeia 1974:960-70

192. Losey, G. S. 1978. The symbiotic be-havior of fishes. In The Behavior ofFishand Other Aquatic Animals, ed. D. I.Mostofsky, pp. 1-31. NY: Academic

193. Losey, G. S. 1979. Fish cleaning symbi-osis: proximate causes of host behavior.Anita. Behav. 27:669-85

194. Lotka, A. J. 1925. Elements of Mathe-matical Biology NY: Dover

195. Lu, K. L., Mesler, M. R. 1981. Ant dis-persal of a neotropical forest floor Ges-neriad. Biotropica 13:159-60

196. Lubchenco, J. 1978. Plant species diver-sity in a marine intertidal community:importance of herbivore food prefer-ence and algal competitive abilities. Am.Nat. 112:23-29

197. Lumpkin, T. A., Plueknett, D. L. 1980.Azolla: botany, physiology, and use as agreen manure. Econ. Bot. 34:111-53

198. MacArthur, R. A. 1972. GeographicalEcology. N-Y: Harper & Row

199. MacArthur, R. A. 1972. Strong, orweak, interactions? In Growth by Intus-susception, ed. E. S. Deevey. Trans.Conn. Acad. Arts Scl. 44:177-88

200. McKey, D. 1975. The ecology of co-evolved seed dispersal systems. In Co-evolution of Animals and Plants, ed. L.E. Gilbert, P. H. Raven, pp. 159-91.Austin, TX: Univ. Texas Press

201. McLaughlin, J. J. A., Zahl, P. A. 1966.Endozoic algae. In Symbiosis, ed. S. M.Henry, Vol. 1. NY: Academic

202. MeNaughton, S. J. 1976. Serengeti mi-gratory wildebeest: facilitation of en-ergy flow by grazing. Science 191:92-94

203. Macior, L. M. 1971. Co-evolution ofplants and animals systematic insightsfrom plant-insect interactions. Taxon20:17-28

204. Malloch, D. W., Pirozynski, K. A.,Raven, P. H. 1980. Ecological and evo-lutionary significance of myeorrhizalsymbioses in vascular plants (a review).Proc. Natl. Acad. Sc£ USA 77:2113-18

205. Margulis, L. 1970. Origin of EukaryoticCellz New Haven, CT: Yale Univ.Press


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



206. Margulis, L. 1976. Genetic and evolu-tionary consequences of symbiosis. Exp.Parasitol. 39:277-349

207.Margulis, L. 1981. Symbiosis in CellEvolution. San Francisco: Freeman.419 pp.

208.Mariscal, R. N. 1970. The nature of thesymbiosis between Indo-Pacific ane-mone fishes and sea anemones. Mar.Biol, 6:58-65

209.Marler, P. R. 1965. Communication inmonkeys and apes. In Primate Behavior:Field Studies of Monkeys and Apes, ed.I. DeVore.

210.Maser, C., Trappe, J. M., Nussbaum, R.A. 1978. Ftmgal-small mammal inter-relationships with emphasis on Oregonconiferous forests. Ecology 59:799-809

211.May, R. M. 1973. Qualitative stabilityin model ecosystems. Ecology 54:638-41

212.May, R. M. 1973. StabiliO~ and Com-plexity in Model Ecosystems. Princeton,NJ: Princeton Univ. Press

213.May, R. M. 1976. (2nd ed., 1981). Mod-els for two interacting populations. InTheoretical Ecology: Principles and Ap-plications, ed. R. M. May, pp. 47-71.Philaddphia: W. B. Saunders

214.Messina, F. J. 1981. Plant protection asa consequence of an ant-membracidmutualism. Ecology 62:1433-40

215.Meyer, H., Provasoli, L., Meyer, F.1979. Lipid biosynthesis in the marineflatworm Convoluta roscoffensis and itsalgal symbiont Platymonas convoluta.Biochim- Biophyx dcta 573:464-80

216.Millbank, J. W., Kershaw, K. A. 1969.Nitrogen metabolism in the cephalodiaof Peltigera aphthosg New PhytoL68:721-29

217.Morse, D. H. 1970. Ecological aspectsof some mixed species foraging flocks ofbirds. EcoL Monogr. 40:119-68

218.Morse, D. H. 1977. Feeding behaviorand predator avoidance in heteros-pecific groups. BioScience 27:332-39

219.Mosquin, T. 1971. Competition for pol-linators as a stimulus for the evolutionof flowering time. Oikos 22:398-402

220.Moynihan, M. H. 1962. The organiza-tion and probable evolution of somemixed species flocks on Neotropiealbirds. Smithsor~ Misc. Coll, 143(7)

221.Moynihan, M. H. 1968. Social mimicry:character convergence versus characterdisplacement. Evolution 22:315-31

222.Muscatine, L. 1974. Endosymbiosis ofcnidarians and algae. In CoelenterateBiology, ed. L. Muscatine, H. M. Len-hoff, pp. 359-95. NY: Academic

223. Muscatine, L., Pool, R. R. 1979. Regu-lation of numbers of intraedlular algae.Proc. R. Soc. Lond. Set, B 204:131-39

224.Muscatine, L., Porter, J. W. 1977. Reefcorals: mutualistie symbiosis adapted tonutrient-poor environments. BioScience27:454-60

225.Nadkarnia, N. M. 1981. Canopy roots:convergent evolution in rainforest nu-trient cycles. Science 214:1023-24

226.Nealson, K. H., Hastings, J. R. 1979.Bacterial bioluminescence: its controland ecological significance. Microbio£Rev. 43:496-518

227.Neyra, C. A., Dobereiner, J. 1977. Ni-trogen fixation in grasses. Adv. dgron.29:1-38

228.Nutman, P. S. 1963. Factors influencingthe balance of mutual advantage inlegume symbiosis. In Symbiotic.4ssocia-tions, ed. P. S. Nutman, B. Mosse.Cambridge: Cambridge Univ. Press

229.O’Dowd, D. J. 1979. Foliar nectar pro-duction and ant activity on a neotropi-cal tree, Ochroma pyramidale,Oecologia 43:233-48

230.O’Dowd, D. J. 1980. Pearl bodies of aneotropieal tree, Ochroma pyramidale:ecological implications. Am. J. Bot.67:543-49

231.O’Dowd, D. J., Hay, M. E. 1980. Mutu-alism between harvester ants and adesert ephemeral: seed escape from ro-dents. Ecology 61:531-46

232.Osman, R. W., Haughsness, J. A. 1981.Mutualism among sessile invertebrates:a mediator of competition and preda-tion. Science 211:846-48

233.Deleted in proof234.Owadally, A. W. 1979. The dodo and

the tambalacoque tree. Science 203:1363~54

235.Owen, D. F. 1980. How plants mayben- .efit from the animals that eat them.Oikos 35:230-35

236.Owen, D. F., Wiegert, R. G. 1976. Doconsumers maximize plant fitness?Oikos 27:488-92

237.Paine, R. T. 1966. Food web complexityand species diversity. Am. Nat.100:65-75

238.Pearse, V. B., Muscatine, L. 1971. Roleof symbiotic algae (zooxanthellae) coral calcification. Biol, Bull, 141:350-63

239. Picket, C. H., Clark, W. 1979. The ex-trafloral nectaries of Opuntia acan-thocarpa (Cactaeeae). Am. J. Bot.66:18-25

240.Pielou, E. C. 1975 Ecological Diversity.NY: Wiley


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



241. Pierce, N. E., Mead, P. S. 1981. Parasi-toids as selective agents in symbiosis be-tween lycaenid butterfly larvae andants. Science 211:1185-87

242. Pijl, L. van der. 1961. Ecological as-pects of flower evolution. Evolution15:44-59

243. Pijl, L. van der. 1969. Principles of Dis-persal in Higher Plants. NY: Springer.154 pp.

244. Pijl, L. van der, Dodson, C. H. 1966.Orchid Flowers, their Pollination andEvolution. Miami: Univ. Miami Press

245. Pilleri, G., Knuekey, J. 1969. Behaviourpatterns of some Delphinidae observedm the western Mediterranean. Z. Tier-psychol. 26:48-72

246. Pirozynski, K. A., Malloeh, D. W.1975. The origin of land plants: a matterof mycotropism. BioSystems 6:153-64

247. Porter, K. G. 1977. The plant-animalinterface in freshwater ecosystems. Am.ScL 65:159-70

248. Porter, J. W. 1976. Autotrophy, hereto-trophy, and resource partitioning inCaribbean reef-building corals. Am.Nat. 110:731-42

249. Post, W. M. III, Travis, C. C., DeAnge-lis, D. L. 1981. Evolution of mutualismbetween species. In Differential Equa-tions and Applications in Ecology, Epi-demics, and Population Problems, ed.K. L. Cooke, S. Busenberg, pp. 183-201. NY: Academic

250. Pound, R. 1893. Symbiosis and mutual-ism. Am. Nat. 27:509-20

251. Powell, J. A., Maekie, R. A. 1966. Bio-logical interactions of moths and Yuccawhipplei. Univ. Calif. Publ. Entomol.42:1-59

251a. Price, P. W,, Bouton, C. E,, Gross, P.,McPheron, B. A., Thompson, J. N.,Weis, A. E. 1980. Interactions amongthree trophic levels: influence of plantson interactions between insect her-bivores and natural enemies. Ann. Rev.EcoL Syst. 11:41-65

252. Proctor, V. W., Yeo, P. 1973. ThePolli-nation of Flowers. London: Collins

253. Pudlo, R. J., Beattie, A. J., Culver, D.C. 1980. Population consequences ofchanges in an ant-seed mutualism inSanguinaria canadensis. Oecologia146:32-37

254. Putwain, P. D., Harper, J. L. 1970.Studies in the dynamics of plant popula-tions III. The interference of associatedspecies on populations of Rumexaeetosa and Rumex acetosella in grass-land. J. Eco£ 58:251-64

255. Quinlan, R. J., Cherrett, J. M. 1978.Aspects of the symbiosis of the leaf cut-

ting ant .4cromyrmex octostTinosus(Reich) and its food fungus. Ecol. En-tomol. 3:221-30

256. Quispel, A., ed. 1974. The Biologyof Nitrogen Fixation. NY: Elsevier.770 pp.

257. Rabatin, S. C. 1980. The occurrence ofthe vesicular-arbuscular myeorrhizalfungus Glomus tenuis with moss. Myco-logia 72:191-95

258. Rabinowitz, D., Rapp, J. K., Sork, V.L., Rathcke, B. J., Reese, G. A.,Weaver, J. C. 1981. Phenological prop-erties of wind- and insect-pollinatedprairie plants. Ecology 62:49-56

259. Regal, P. J. 1977. Ecology and evolu-tion of flowering plant dominance.Science 196:622-29

260. Rehr, S. S., Feeney, P. O., Janzen, D. H.1973. Chemical defense in CentralAmerican non-ant acacias. J.. Anita.Ecol. 42:405-16

261. Richardson, J. L 1980. The organismiccommunity: resilience of an embattledecological concept. BioScience 30:465-71

262. Rick, C. M., Bowman, R. I. 1961.Galapagos tomatoes and tortoises. Evo-lution 15:407-17

263. Rickson, F. R. 1977. Progressive loss ofant-related traits of Cecropia peltata onselected Caribbean islands:.4m. Z Bot.64:585-92

264. Riseh, S., Boucher, D. 1976. Whatecologists look for. Bull. Ecol. Soc. Am.57(3):8-9

265. Risch, S., McClure, M., Vandermeer,J., Waltz, S. 1977. Mutualism betweenthree species of tropical 15"per (Pipera-ceae) and their ant inhabitants. Am.Midl. Nat. 98:433-44

266. Risch, S. J., Rickson, F. R. 1981. Mutu-alism in which ants must be present be-fore plants produce food bodies. Nature291:149-50

267. Rodgers, G. A., Stewart, W. D. P. 1977.The cyanophyte-hepatic symbiosis. NewPhytol. 78:441-58

268. Deleted in proof269. Ross, D. M. 1974. Evolutionary aspects

of associations between crabs and seaanemones. In Symbiosis in the Sea, ed.W. B. Vernberg. Columbia, SC: Univ.South Carolina Press

270. Ross, G. N. 1966. Life history studieson Mexican butterflies IV. The ecologyand ethology of Anatole rossi, a myr-mecophilous metalmark (Lepidoptera:Riodinidae). Ann. Entomol. Soc. Am.59:985-1004

271. Roughgarden, J. 1975. Evolution ofma-


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



fine symbiosis: a simple cost-benefitmodel. Ecology 56:1201-8

272.Schemske, D. N. 1981. Floral conver-gence and pollinator sharing in two bee-pollinated tropical herbs. Ecology 62:946-54

273.Schneider, A. 1897. The phenomena ofsymbiosis. Mink Bot. Stud. 1:923-48

274,Scott, G. D. 1969. Plant Symbiosi~London: Edward Arnold

275.Scott, G. D. 1973. Evolutionary aspectsof symbiosis. In The Lichens, ed. V. Ah-madjian, M. E. Hale. NY: Academic

276.Scudo, F. M., Ziegier, J. R. 1976. Vladi-mir Aleksandrovich Kostitzin and theo-retical ecology. Theor. Pop. Biol. 10:395-412

277.Scudo, F. M., Ziegler, J. R. 1977. TheGolden Age of Theoretical Ecology."1923-1940. NY: Springer

278.Scifert, R. P., Seifert, F. H. 1976. Acommunity matrix analysis of Heliconiainsect communities. Am. Nat. I10:461-83

279.Deleted in proof280.Sharma, I. 1977. Development of com-

mensalism between prey and predator.Am. Na~ 111:1009-10

281.Simberlotf, D., Brown, B. J., Lowrie, S.1978. Isopod and insect root borers maybenefit Florida mangroves. Science201:630-32

282.Slobodkin, L. B., Fishelson, L. 1974.The effect of the cleaner-fish, Labroidesdimidiatus, on the point diversity offishes on the reef front at Eilat. Am. Nat.108:369-76

283.Smith, D. C. 1979. From extracdlularto intracellular: the establishment ofsymbiosis. Proc. R. Soc. Lond. Ser. B204:115-30

284.Smith, D., Muscatine, L., Lewis, D.1969. Carbohydrate movement fromautotrophs to heterotrophs in mutualis-tie symbioses. Biol. Rev. 44:17-90

285.Smith, N. G. 1968. The advantage ofbeing parasitized. Nature 219:690-94

286.Smith, N. G. 1980. Some evolutionary,ecological, and behavioural correlatesof cominunal nesting by birds withwasps or bees. AcL Congr. lnt. Orni-thol., 17th, pp. 1199-1205

287.Snow, D. W. 1965. A possible selectivefactor in the evolution of fruiting sea-sons in tropical forests. Oikos 15:274-81

288. Soberon, J. M., Martinez del Rio, C.1981. The d]naamies of a plant-pollina-tor interaction. J. Theor. Biol. 91:363-78

289.Start, M. P. 1975. A generalized schemefor classifying orgamsmic associations.In Symbiosis, cd. D. H. Jennings, D. L.

Lee, pp. 1-20. Cambridge: CambridgeUniv. Press

290.Stebbins, G. L. 1957. Self-fertilizationand population variability in higherplants. Am. Nat. 91:337-54

291.Steinhans, E. A. 1946. Insect Microbi-ology. Ithaca, NY: Comstock

292.Stenseth, N. C. 1978. Do grazers maxi-mize individual plant fitness? Oikos31:299-306

293.Sussman, R. W., Raven, P. H. 1978.Pollination by lemurs and marsupials:an archaic eoevolutionary system.Science 200:731-36

294.Taylor, D. L. 1974. Symbiotic marinealgae: taxonomy and biological fitness.In Symbiosis in the Sea, ed. W. B. Vern-berg. Columbia, SC: Univ. SouthCarolina Press

295.Taylor, F. R. J. 1979. Symbionticismrevisited: a discussion of the evolution-ary impact of intracellular symbioses.Proc. R. Soc. Lond. Ser. B. 204:267-86

296.Temple, S. A. 1977. Plant-animal mutu-alism: coevolution with dodo leads tonear extinction of plant. Science197:885-86

297.Thien, L. B. 1980. Patterns of pollina-tion in the primitive angiosperms. Bio-tropica 12:1-13

298.Thompson, J. N. 1981. Elalsomes andfleshy fruits: phenology and selectionpressures for ant-dispersed seeds.Nat. 117:104-8

299.Thompson, J. N., Willson, M. F. 1978.Disturbance and the dispersal of fleshyfruits. Science 200:1161-63

300.Thompson, J. N., Willson, M. F. 1979.Evolution of temperate fruit/bird in-teractions: phenological strategies. Evo-lution 33:973-82

301.Tikhomirov, B. A. 1966. Thelnterrela-tionships of Animal Life and Vegeta-tional Cover of the Tundra. Jerusalem:Israeli Progr. Set. Trans.

302.Tilman, D. 1978. Cherries, ants andtent caterpillars: timing of nectar pro-duction in relation to susceptibility ofcaterpillars to ant predation. Ecology59:686-92

303.Torrey, J. G., Tjepkma, J. D., eds. 1979.Symbiotic nitrogen fixation in ac-tinomycete-nodulated plants. Bot. Gaz.140:Suppl. Sl-S146

304.Trager, W. 1970. Symbiosis~ NY: VanNostrand

305.Travis, C. C., Post, W. M. III. 1979.Dynamics and comparative statics ofmutualistie communities. J. Theor. Biol.78:553-71

306.Travis, C. C., Post, W. M. III, DeAnge-Its, D. L., Perkowski, J. 1980. Analysis


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.



of compensatory Leslie matrix modelsfor competing species. Theor. Pop. Biol.18:16-30

307. Trench, R. K. 1979. The cell biology ofplant animal symbiosiS. Ann. Rev. PlantPhysiol. 30:485-531

308. Trinick, M. J. 1973. Symbiosis betweenRhizobium and the nonqegume Tremaasperg Nature 244:459-60

309. Trinick, M. J. 1980. Growth of Para-sponia in agar tube culture and symbi-otic effectiveness of isolates from Para-sponia spp. New Phytol. 85:37-45

310. Valerio, C. E. 1975. A unique case ofmutualism. Am. Nat. 109:235-38

311. van Beneden, P. J. 1875. Les Commen-saux et les Parasite~ Paris: Biblio.Sei. Int.

312. Van Valen, L. 1965. Morphologicalvariation and width of ecological niche.Ara. Nat. 99:377-90

313. Vance, R. R. 1978. A mutualistie in-teraction between a sessile marine clamand its epibionts. Ecology 59:679-85

314. Vandermeer, J. H. 1980. Indirect mutu-alism: variations on a theme by StephenLevine. Am. Nat. 116:441-48

315. Vandermeer, J. H., Boueher, D. H.1978. Varieties of mutualistie interac-tions in population models. J. Theor.Biol 74:549-58

316. Viera da Silva, J. 1979. Introduction a laTheorie Ecologique. Paris: Masson

317. Vine, I. 1971. Risk of visual detectionand pursuit by a predator and the selec-tive advantage of flocking behavior, £Theor. Biol. 30:405-22

318. Volterra, V. 1936. Variations and fluc-tuations of the number of individuals ofanimal species living together. In TheGolden Age of Theoretical Ecology:1923-1940, ed. F. M. Scudo, J. R. Zie-gler. NY: Springer

319. Waser, N. M. 1978. Competition forhummingbird pollination and sequen-tial flowering in two Colorado wildflow-ers. Ecology 59:934-44

320. Waser, N. M., Real, U A. 1979. Effec-tive mutualism between sequentiallyflowering plant species. Nature 281:670-72

321. Washburn, S. L., DeVore, I. 1961. Thesocial life of baboons. ScL Am.204:67-71

322. Watt, W. B., Hoch, P. C., Mills, S. G.1974. Nectar resource use by Coliasbutterflies. Oecologia 14:353-74

323. Way, M. J. 1963. Mutualism betweenants and honeydew-producing Homopt-era. dng Rev. Entomol. 8:307--44

324. Weis, D. S. 1977. Synchronous devdop-ment of symbiotic chlorellae withinParamecium bursaria. Trans..4m. Mi-

croscL Soc. 96:82-86325. West, H. H., Harrigan, J. 1979. Symbio-

sis and development in two populationsof Elysia chlorotica. Am. J. ZooL19:958

326. Whatley, J. M., John, P., Whatley, F.R. 1979. From extracellular to intracel-lular: the establishment of mito-chondria and chloroplasts. Proc. R. Soc.Lond. Set. B 204:165-87

327. Wheeler, W. M. 1928. The Social In-sects," Their Origin and Evolution. Lon-don: Kegan Paul, Trench, Trubner andCo.

328. Wheelwright, N. T., Orians, G. H.1982. Seed dispersal by animals: con-trasts with pollen dispersal, problems ofterminology, and constraints on coevo-hition. Am. Nat. 119:402-13

329. Whittaker, R. H. 1975. Communitiesand Ecosystems. N’Y: Macmillan. 2nded.

330. Whittaker, R. H., Levin, S. A., eds.1975. Niche: Theory and ~4pplicatiogStroudsberg, PA: Halstead Press

331. Wiens, D., Rourke, J. P. 1978. Rodentpollination in southern African Proteaspp. Nature 276:71-73

332. Wilkerson, F. P. 1980. Bacterial sym-bionts on green hydra and their effect onphosphate uptake. Microbial Ecol.6:85-92

333. Wilkinson, C. R., Fay, P. 1979. Nitro-gen fixation in coral reef sponges withsymbiotic cyanobacteria. Nature279:527-29

334. Williams, G. C. 1966. Adaptation andNatural Selection. Princeton, N J:Princeton Univ. Press

335. Wilson, D. S. 1976. Evolution on thelevel of communities. Science192:1358-60

336. Wilson, D. S. 1980. The NaturalSelec-tion of Populations and Communities.Menlo Park, CA: Benjamin/CummingsPubl. Co.

337. Wilson, E. O. 1971. The Insect Societies.Cambridge, MA: Belknap Press

338. Wilson, E. O. 1975. Sociobiology. Cam-bridge, MA: Belknap Press

339. Wilson, P. W. 1940. The Biochemistryof Symbiotic Nitrogen Fixatiog Madi-son, WI: Univ. Wisconsin Press.

340. Worster, D. 1977. Nature’s Economy:the Roots of Ecology. San Francisco:Sierra Club Books

341. Yeoh, H. T., Bungay, H. R., Krieg, N.R. 1968. A microbial interaction involv-ing combined mutualism and inhibition.Can. J. Microbiol. 14:491-92

342. Zimmerman, J. 1932. 0ber die Ex-trattorale Nectarien der Angiosperrnen.Bot. Centr. Beih. 49:99-196


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.


Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.

Ann

u. R

ev. E

col.

Syst

. 198

2.13

:315

-347

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Uni

vers

ity o

f K

anas

-Law

renc

e &

Edw

ards

on

09/2

6/05

. For

per

sona

l use

onl

y.

Documents

Boucher Ecology of Mutualism